A retracted article can continue to be cited for years after its withdrawal.
The database flags it. The DOI remains unchanged. Yet five years later, the article may still be cited dozens of times in publications that never checked its status. This is not a marginal issue. It is one of the most persistent blind spots in scientific publishing. Above all, it directly affects the credibility of journals that unknowingly publish these citations.
For editors-in-chief and editorial teams at STM journals (scientific, technical and medical), this risk is particularly insidious. Indeed, it may not be visible at the time of publication. Instead, it can quietly accumulate in the literature until a reader, journalist or ethics committee eventually identifies it.
A study published in Quantitative Science Studies (MIT Press) by Tzu-Kun Hsiao and Jodi Schneider confirms that retraction does not stop an article from circulating. The authors analyzed 13,252 citation contexts published after retraction. Only 722 of them, or 5.4%, explicitly mentioned that the cited article had been retracted. They also refer to an earlier study of 460 retracted genetics papers, which found that 23% of their citations had been published after retraction. In other words, in the vast majority of the cases analyzed, the retracted status of the article was not disclosed in the citation context. ( Tzu-Kun Hsiao and Jodi Schneider, Quantitative Science Studies)
One case illustrates just how long the problem can persist. Researcher Jodi Schneider documented a fraudulent clinical trial that was still being cited eleven years after its retraction, with citations increasing after withdrawal rather than decreasing (University of Illinois News Bureau).
Why Retracted Articles Continue to Be Cited
A retraction signals that an article should no longer be treated as a reliable scientific source without taking the retraction notice into account. Yet retraction rarely stops its bibliographic circulation. Several factors explain why a retracted article may continue to be cited.
• Retraction metadata is fragmented.
Retraction information may be distributed across several sources and may not appear in the same way depending on the tools being used. As a result, the original publisher may flag an article as retracted while other databases used by researchers do not immediately display that information. This fragmentation therefore makes reference checking more difficult.
• Reference management software does not always make retraction status sufficiently visible.
Zotero, EndNote and Mendeley allow researchers to import a DOI and its metadata. However, these tools do not necessarily ensure that an author is clearly informed of the retraction status of a reference. Consequently, a researcher may cite a retracted article without receiving a sufficiently visible warning.
• Peer review does not always cover this issue.
Reviewers primarily assess the scientific relevance, methodology and results of a submitted article. By contrast, they rarely check the retraction status of every bibliographic reference. Yet a research paper may contain dozens of references. As a result, editorial teams may have to spend significant time checking references manually, DOI by DOI.
• Citation cascades can perpetuate the problem.
Citing a retracted article is not necessarily problematic in itself. A citation can be entirely legitimate when it explicitly mentions the retraction or discusses the case in question.
However, citing a retracted article as a valid scientific source without disclosing its status can create concrete risks for a journal.
A Risk That Goes Beyond a Simple Bibliographic Error
Citing a retracted article is not merely a formatting or reference issue. It can have concrete consequences for a scientific journal.
- Scientific risk: if the conclusions of a new paper rely, even partially, on data or methods invalidated by the retraction, the validity of the new publication may itself be called into question.
- Reputational risk: citations of retracted articles are increasingly documented publicly, particularly through Retraction Watch. They may also be reported in the scientific press or discussed across professional networks.
- Ethical and editorial risk: for journals that follow COPE recommendations, the use of retracted references without appropriate context may reveal weaknesses in editorial quality-control processes.
- Cumulative impact on bibliometric indicators: citations of retracted articles continue to feed bibliometric data and may artificially maintain the visibility of invalidated research.
For independent journals and smaller publishing organizations, this issue can be even more difficult to manage. These teams often have fewer resources than large publishing groups to implement systematic bibliographic reference checking.
Why Manual Reference Checking Reaches Its Limits
A scientific article often contains between 30 and 80 bibliographic references, and sometimes many more.
Checking every reference for a possible retraction therefore represents a significant workload. For each reference, the editorial team may need to check the DOI, consult several databases and identify any relevant retraction notice. When multiplied by the number of manuscripts processed each year, this task becomes difficult to perform systematically.
As a result, the process often relies on the individual vigilance of authors, reviewers or editorial staff. This gap between editorial expectations and actual operational capacity is precisely what allows the risk to persist. Ideally, every reference should be checked. In practice, editorial teams rarely have enough time to perform this process manually.
Automating the Detection of Retracted Articles
Automation provides a structural response to this problem.
This is precisely where LibScience is positioned: automating bibliographic reference checking, including the cross-checking of references against retraction databases such as Retraction Watch and Crossref.
In practical terms, the solution allows editorial teams to:
- automatically identify references pointing to a retracted article;
- perform this check during submission or peer review;
- receive a clear and documented alert;
- avoid manually checking every DOI against several external databases;
- integrate reference verification as a systematic step in the editorial workflow.
The goal is not to replace editorial judgment. Rather, it is to give editorial teams visibility into a risk that is otherwise difficult to detect manually.
The editor can then decide whether the citation should be removed, replaced or retained with an explicit mention of the retraction.
What Journals Can Do Right Now
While waiting for, or in addition to, an automated solution, several practices can help reduce the risk of citing retracted articles.
- Raise author awareness in the author guidelines and ask researchers to check the status of their references before submission.
- Add an explicit checkpoint to the editorial review checklist.
- Consult the Retraction Watch Database for the most frequently cited or oldest references in a submitted manuscript.
- Check DOIs for key references in order to identify possible updates or retractions.
- Document the process so that the journal can demonstrate, if questions arise later, that reference verification is part of its editorial workflow.
These measures reduce exposure to the risk. However, they do not eliminate it completely. By nature, manual verification remains difficult to apply systematically given the current volume of scientific publications.
In Summary
The risk associated with citing a retracted article is invisible because it does not always appear at the time of publication. It may only become apparent years later, after the article has already been read, cited and integrated into the scientific literature.
For independent and institutional STM journals alike, checking the status of references is therefore becoming an important issue of scientific integrity. Introducing systematic checks, whether manual or automated, can reduce this risk and strengthen the reliability of the editorial process.
In this respect, checking for retracted articles is increasingly becoming an editorial quality-control step in its own right, alongside plagiarism detection and conflict-of-interest management.
LibScience develops an automated bibliographic reference verification solution for scientific journals and publishers, including the detection of retracted articles, automatic DOI enrichment and contextual citation relevance checking. Discover LibScience