When a research paper is retracted, it is officially withdrawn from the scientific record because its findings cannot be trusted. You might expect researchers to stop citing it immediately. Yet study after study has shown that retracted papers often continue to be cited for years, frequently without any mention that they were retracted.
This article explores why retracted papers still get cited, what the evidence shows, why the problem matters, and what researchers, journals, and technology providers can do to stop unreliable work from spreading.
Table of Contents
What Happens When a Paper Is Retracted?
When a journal retracts an article, it usually publishes a retraction notice and marks the original paper as retracted on its website. The paper often remains available so readers can see what happened. However, a retraction notice does not automatically remove the paper from databases, reference lists, downloaded files, or the memories of researchers who have already read it.
What the Evidence Shows
Researchers who study citation patterns have documented the problem in detail.
| Finding | What It Shows |
|---|---|
| Only about 5 percent of post-retraction citations in one large analysis acknowledged the retraction | Most citing authors do not mention retracted status |
| A retracted 2005 medical paper was still being cited more than a decade later | Retracted work can circulate for many years |
| In that case, 96 percent of studied post-retraction citations did not mention the retraction | Many citing authors appear unaware |
| Only about 3 percent of retracted papers were flagged in all four major databases examined | Retraction information is inconsistent across systems |
A Case Study
Researchers at the University of Illinois examined a 2005 respiratory medicine paper reporting that omega-3 fatty acids reduced inflammation in patients with chronic obstructive pulmonary disease. The paper was retracted in 2008 for falsified clinical trial data. Yet it continued to be cited for more than 11 years. Of the post-retraction citations whose context could be analysed, 96 percent did not mention the retraction, and many described the paper’s methods and results in detail as if they were valid.
Recent Retractions
A 2025 analysis of 152 recent retractions found that about 30 percent gained new citations after their retraction date. Citations did not appear to slow after retraction notices were published.
Why Retracted Papers Keep Getting Cited
1. Publication Delays
Writing, reviewing, and publishing a paper can take many months. An author may cite a valid paper when drafting a manuscript, but by the time their own work is published, the cited paper may have been retracted. Studies of biomedical publishing have found submission-to-publication times ranging from a few months to well over a year.
2. Hard to Change References Late
Once a manuscript has been accepted, removing or replacing a reference may require rewriting parts of the paper. Authors and journals may not have processes to catch these late changes.
3. Inconsistent Database Labelling
Different academic databases and indexes do not always flag retracted papers consistently. A paper may appear as retracted in one database but not in another, so researchers relying on a single source may miss the warning.
4. Downloaded Copies and Reference Libraries
Researchers often save PDFs and build personal reference libraries. Older downloaded versions do not update when a paper is retracted, so authors may continue citing work they saved years earlier.
5. Citing Without Reading the Original
Some authors cite papers based on other articles, reviews, or textbooks rather than checking the original source. If those secondary sources do not mention the retraction, the error spreads.
6. Lack of Awareness
Researchers may not routinely check whether their references have been retracted, and some may not know how to do so.
7. Legitimate Citations
Not every post-retraction citation is a mistake. Some authors cite retracted papers to discuss research misconduct, explain why certain findings were rejected, or study retractions themselves. The key is to clearly state that the paper has been retracted.
Why This Matters
- Spreading false information: unreliable findings continue to influence new research
- Distorting reviews: systematic reviews and meta-analyses may include invalid studies
- Health risks: in medicine, retracted findings could influence clinical decisions
- Wasted resources: researchers may build experiments on faulty foundations
- Undermining trust: continued citation weakens the purpose of retractions
Solutions Being Developed
Better Retraction Data
In 2023, Crossref acquired the Retraction Watch database and made its retraction data openly available. This makes it easier for publishers, databases, and software tools to identify retracted papers automatically.
Reference Screening by Journals
Experts suggest that journals screen reference lists for retracted papers at submission, revision, and acceptance. Finding a retracted citation late in the process could trigger a required correction before publication.
Alerts for Authors
Systems could notify authors automatically when a paper they have cited, or are about to cite, is retracted, allowing them to update their work.
Improved Labelling
Clear, consistent retraction labels across publisher websites, databases, search engines, and PDF files help readers recognise retracted work wherever they find it.
Reference Manager Tools
Some reference management tools can flag retracted items in a researcher’s library, helping authors catch problems before submission.
The Ripple Effect of Citations
The problem does not stop with direct citations. When a new paper cites a retracted study without noting its status, other researchers may later cite that new paper, spreading the unreliable claim further. In the case study above, researchers also identified thousands of second-generation citations, showing how one retracted finding can echo through the literature. Review articles and textbooks can extend this reach even further, because many readers trust them as summaries of established knowledge.
Not Every Citation Is Equal
Researchers studying this issue often look at how a retracted paper is cited. Some citations simply list it in passing, while others rely on its methods or results as evidence. Citations that use retracted findings to support new conclusions are the most concerning, since they can directly influence research directions and decisions.
What Researchers Can Do
- Check reference lists for retractions before submitting and again before final acceptance
- Use databases and tools that display retraction information
- Read and cite original sources rather than relying on secondary summaries
- Update personal reference libraries regularly
- If citing a retracted paper intentionally, state its retracted status clearly
- Contact the journal if you discover you cited a retracted paper after publication
What Journals and Publishers Can Do
- Screen references automatically for retracted papers
- Label retracted articles clearly on all versions, including PDFs
- Share retraction information promptly with databases and indexing services
- Publish clear retraction notices that explain the reasons
- Link retraction notices and original articles in metadata
Final Thoughts
Understanding why retracted papers still get cited reveals a gap between correcting the scientific record and making sure everyone notices. Publication delays, inconsistent databases, saved PDFs, and lack of routine checks all allow retracted work to keep circulating. With open retraction data, automated reference screening, better labelling, and careful habits by researchers, the research community can help ensure that once a paper is retracted, it stops shaping new science.
