Why Retracted Papers Keep Getting Cited

A retraction is supposed to warn readers that a paper's conclusions cannot be trusted, but retracted articles keep accumulating citations, often with no sign that they were ever withdrawn. The warning fails to travel because copies, databases, and reference managers do not always carry the retraction label, so the fix has to be actively checked rather than assumed.

What a retraction is supposed to do

A retraction is the strongest correction the scholarly system has. It declares that a paper's central findings can no longer be trusted, whether because of an honest error or misconduct, and it warns everyone downstream to stop relying on the work. Publication ethics guidance is clear that the point is to protect the integrity of the record, not to punish the authors.

For that warning to work, it has to reach the people who might use the paper. A retraction that no one sees is a warning that never arrives.

The zombie-citation problem

In practice, retracted papers keep getting cited, often long after they are withdrawn and usually as if nothing had happened. A bibliometric study of primary care journals found that retracted articles continued to be cited heavily, with the large majority of their citations arriving after the retraction, and each retracted paper collecting a substantial number of them.

These are sometimes called zombie citations: references to work that is formally dead but still circulating. The danger is that a retracted result quietly seeds later reviews, recommendations, and studies, so a single discredited paper keeps shaping a field through everything built on top of it.

Why the warning does not travel

The retraction lives in one place, but the paper lives in many. Copies sit in personal folders, on co-author websites, in reference managers, and in third party databases, and those copies rarely update themselves when the original is flagged. Someone who saved or cited the paper before the retraction may never learn that anything changed.

There is also a quieter reason. Authors often cite from their own past reading and existing reference lists, and few re-check every source at the moment of writing. Unless a database actively surfaces the retraction at the point of use, the flag is easy to miss even for a careful person.

What the standards require

The standards are not vague about this. Guidance from the Committee on Publication Ethics states that a retracted article should be identified unmistakably as retracted in every online location, linked in both directions to its notice, and labeled so that bibliographic databases carry the flag wherever the paper appears. The aim is that anyone who finds the paper also finds out it was retracted.

The gap is between the standard and its reach. Journals and major indexes have improved at labeling, but the label does not always propagate to every copy and every tool a reader might use, which is where zombie citations slip through.

Why continued citation matters

In fields that touch patient care, a lingering retracted result is not just an academic tidiness problem. If a withdrawn finding continues to be cited into reviews and recommendations, it can influence how evidence is summarized and, indirectly, how clinicians and patients understand a question. The original correction was meant to prevent exactly that.

This is the practical argument for taking retraction seriously as readers, not only as editors. A retraction only protects the record if the people building on the record actually act on it.

How to check a reference before you trust it

The habit that defends against zombie citations is simple to state. Before leaning on a key paper, check its current status in a source that tracks retractions, and look for a retraction or concern notice attached to the record. Reference managers and databases increasingly flag these, and dedicated retraction databases exist for exactly this purpose.

None of this requires suspicion of every citation. It requires a quick check on the ones that carry weight in your reasoning. The correction has already been issued; the remaining work is to make sure it reaches you before you pass the finding along.