03 · What You Need to Know
What Should You Do With a Retracted Study You Have Already Found?
Start by Confirming That the Paper Really Has Been Retracted
Do not act on an informal allegation, social-media discussion, or secondhand claim alone. Verify the article's current status through the publisher's record and the formal retraction notice. Depending on the discipline, linked records in PubMed, Crossref metadata, and Retraction Watch data can provide additional verification.
PubMed explicitly links retracted publications with their retraction notices for records it indexes. The National Library of Medicine assigns Retracted Publication to the affected article and Retraction Notice to the corresponding notice. Crossref also makes retraction metadata available through its services, including data supplied by Retraction Watch.
If you have not yet verified the status, first determine whether the paper has actually been corrected, retracted, or otherwise formally updated.
Then Read the Retraction Notice
The word retracted tells you that a formal action occurred. It does not tell you everything you need to know about what happened.
Retractions occur for different reasons. COPE's retraction guidance includes circumstances in which findings are considered unreliable because of major error or misconduct, as well as other serious problems involving publication ethics. The National Library of Medicine likewise notes that it does not distinguish in its retraction indexing between articles retracted because of honest error and those retracted because of scientific misconduct or plagiarism.
The retraction notice is therefore essential. Ask what the journal says is unreliable, why the publication was retracted, whether the entire paper is affected, and whether a corrected or republished version exists.
Watch Out
Do not translate “retracted” into “fraudulent” unless an authoritative source actually establishes misconduct. Retraction concerns the status of the publication; the cause and responsibility require separate evidence.
Separate the Paper From the Claim You Were Using It to Support
Suppose your literature review currently says that an intervention improves student achievement and cites five studies, one of which has subsequently been retracted because its findings were judged unreliable.
Your immediate question should not be “Can I keep this citation?” Ask instead: “What evidence now supports this claim?”
If the retracted study was one of several independent sources pointing in the same direction, removing it may leave the broader conclusion substantially unchanged. If it was the only source supporting the claim, the claim itself may need to be removed or substantially qualified.
If it was the largest or most influential study in the evidence base, you may need to reassess how the remaining literature should be interpreted even when several other papers exist.
This is part of determining what to do when an important paper develops serious problems, rather than treating reference management as the entire issue.
Do Not Cite the Retracted Finding as If Nothing Happened
This is the central practical rule.
If a paper has been retracted because the findings you intended to use are no longer considered reliable, it should not continue appearing alongside unretracted studies as though all of the citations provide equivalent evidential support.
For example, a sentence such as “Several studies demonstrate that the intervention improves retention” followed by citations that silently include a retracted study can mislead readers. The problem is not merely formatting. The citation communicates evidential support that the publication no longer legitimately provides in the same way.
COPE's retraction guidance emphasizes that retractions should occur promptly in part to minimize researchers continuing to cite erroneous work, act on its findings, or allow unreliable publications to distort evidence syntheses.
A Retracted Paper Can Still Be Relevant as an Object of Study
Retraction does not erase the historical existence of a publication. Sometimes that history is exactly what you need to discuss.
A retracted paper may remain relevant when your literature review examines research misconduct, publication ethics, scientific self-correction, misinformation, citation behavior after retraction, the development of a disputed theory, or the historical influence of a particular finding.
In those situations, the paper is not being cited to establish that its original scientific claim is true. It is being cited because the publication, its influence, or its retraction is itself part of the phenomenon under examination.
Citing as evidence
The retracted paper is presented as support for the substantive research claim originally reported in the article.
Citing as an object of discussion
The paper is cited because its publication, influence, correction, retraction, or consequences are relevant to the scholarly argument.
This distinction should be obvious to the reader from the sentence itself, not hidden in a footnote that few people will notice.
Historical Accuracy May Require Mentioning a Retracted Study
Imagine reviewing the development of a theory that became influential partly because of one highly cited study. That study was later retracted.
Simply deleting the study from the historical narrative could create a misleading account of how the field developed. A more accurate review might explain that the publication initially influenced subsequent work but was later retracted and should no longer be treated as reliable support for the original claim.
This approach preserves two truths simultaneously: the paper influenced the literature, and its current evidential status has changed.
Be Explicit About the Retraction in the Prose
If a retracted article genuinely needs to remain in your literature review, do not rely solely on the reference list to communicate its status.
State clearly in the surrounding prose that the article was retracted. Depending on why it is being discussed, you may also need to identify the reason reported in the retraction notice and explain how the retraction affects interpretation.
For example, instead of writing:
“Early research reported a substantial association between X and Y.”
A more transparent formulation would be:
“An early study reported a substantial association between X and Y, but the article was subsequently retracted after the journal determined that the affected findings were unreliable.”
The exact wording should reflect the formal notice. Do not add accusations or causal explanations that the notice does not support.
Cite the Retraction Notice When It Matters to Your Claim
If your sentence discusses the fact, reason, scope, or circumstances of a retraction, the retraction notice itself is an important source.
The National Library of Medicine treats the retracted publication and the retraction notice as distinct but linked publication types. Its citation guidance also provides conventions for identifying that an article has a subsequent retraction notice.
This distinction is useful conceptually even outside biomedical citation styles. The original paper documents what was published; the retraction notice documents the formal action taken afterward.
Your journal or citation style may have specific conventions for formatting retracted publications and notices, so check the applicable author guidelines rather than inventing your own notation.
Do Not Assume Every Sentence in a Retracted Paper Is False
A retraction means the publication should no longer be relied upon in its original form, but it does not logically establish that every sentence, citation, background statement, or observation in the document is false.
This distinction should not be used as an excuse to salvage the paper's central findings when the retraction has undermined them. It matters mainly because retractions vary in scope and cause.
If you believe a particular unaffected element remains important, the safer approach is usually to locate and cite an independent, reliable source for that information rather than relying unnecessarily on the retracted publication.
Replace the Evidence When You Can
If you originally cited a retracted article for a routine factual or empirical claim, search for independent evidence addressing the same point.
Perhaps the claim is supported by other primary studies. Perhaps a later replication exists. A systematic review may provide a better overview. Alternatively, the remaining literature may reveal that the supposedly established finding was much less secure than the retracted paper made it appear.
Replacing the citation should not mean finding any paper that reaches the same conclusion. Critically evaluate the replacement evidence on its own merits. Otherwise, you risk preserving a preferred claim while merely changing the reference attached to it.
Retraction Can Change a Systematic Review or Meta-Analysis
The consequences become especially important when a retracted study has entered an evidence synthesis.
COPE specifically notes the danger of erroneous publications contributing to incorrect conclusions through mechanisms such as double counting in meta-analysis. If an included study is subsequently retracted, reviewers should determine whether the synthesis and its conclusions remain valid without that study.
The appropriate response depends on the study's contribution. Removing a small, low-weight study may barely change a pooled estimate. Removing a large or influential study could materially alter the effect estimate, heterogeneity, certainty assessment, or interpretation.
When evaluating a systematic review or meta-analysis, checking whether important included studies have subsequently been retracted can therefore be consequential, particularly for older reviews.
A Retraction May Also Affect Papers That Cite the Retracted Study
Discovering one retracted article sometimes creates a broader problem. Other papers may have relied heavily on its findings, reused its data, incorporated its assumptions, or built later analyses upon it.
Those later papers do not automatically become invalid. Their dependence on the retracted work needs to be examined.
If a later study merely mentioned the retracted paper in its introduction, the consequences may be negligible. If its model, intervention, dataset, or central premise depends directly on the invalidated work, the implications could be substantial.
This is why a literature review should treat evidence as a network of claims and dependencies rather than as a stack of independent citations.
Retraction Does Not Automatically Require Deleting the Entire Topic From Your Review
Suppose an influential paper supporting a theory is retracted. That does not establish that the theory itself is false. Other independent evidence may still support it.
Likewise, a retracted intervention study does not prove that the intervention is ineffective. It means that the affected study should no longer carry the evidential weight you previously assigned to it.
Re-evaluate the broader evidence rather than reversing the conclusion mechanically. This is the same principle involved in distinguishing a problem with one piece of research from a problem fatal to a broader inference.
Retraction Status Can Change After Your Literature Search
A paper may be unretracted when you first collect it and retracted months or years later. This matters for theses, systematic reviews, books, guidelines, and other projects with long development periods.
Crossref currently incorporates Retraction Watch retractions into its production metadata services, and the Retraction Watch dataset is updated on working days. PubMed likewise maintains linked records for retractions and other post-publication updates in the literature it covers.
For a consequential evidence synthesis, it may therefore be prudent to recheck the publication status of included or heavily relied-upon studies near the end of the project rather than assuming that the status recorded during the original search remains current.
