Retraction is not the same as disagreement
A paper is retracted when something is wrong with it that cannot be repaired: fabricated data, an error invalidating the results, undisclosed conflicts, or duplicate publication. That is a different matter from a study whose findings were later contradicted, which is ordinary science.
The distinction matters for how you treat it. A contradicted study can be cited and discussed. A retracted one cannot be cited as evidence at all, whatever it says.
How to check, quickly
Open the source at the publisher rather than reading a downloaded copy, because that is where the notice appears. Publishers mark retracted articles clearly, usually with a banner and a linked notice.
The downloaded PDF frequently carries no marking whatever, which is exactly why a file saved a year ago is the most dangerous version of a source you can be working from.
Watch for the expression of concern
Between publication and retraction sits an intermediate state, where a journal has flagged a paper as under investigation without withdrawing it. That is a signal to find something else rather than to proceed carefully.
It also tends to be invisible unless you look at the publisher's record, since databases update at different rates and some will show the original entry untouched for months.
Check what a paper's citations are saying
Where a study is central to your argument, look at what has cited it recently. A cluster of recent citations that are all critical is worth reading before you rely on it, and it occasionally surfaces problems short of retraction.
This is the same check that tells you whether an older source has aged well, so it does double work on anything your argument leans on heavily.
What to do if you have already cited one
Remove it and find a replacement, which is usually straightforward because a retracted finding was either never real or has been established properly elsewhere. If the claim cannot be supported without it, the claim needs to change.
Do not simply move it to a footnote or soften the sentence around it. A retracted source supporting a softened claim is still a retracted source supporting a claim.
Why this one is unforgiving
Most citation errors read as carelessness. A retracted source reads as not having checked at all, and in health and social sciences the well-known retractions are known to the people marking your work.
That is the whole reason a minute per source is worth spending. It is the cheapest possible insurance against the one error that colours a reader's view of everything else you wrote.
Where retracted work keeps circulating
Reference lists are the main route. A retracted paper cited in a review published before the retraction stays there permanently, and anybody working from that review inherits the citation without ever seeing the notice.
This is the practical argument for following a citation trail back to its source rather than trusting a secondary account of it. The check that protects you against claim drift protects you against this as well.
It is also why a source inherited from a reading list deserves the same check as one you found yourself. A list assembled three years ago reflects what was known three years ago, and nobody goes back to revise it.
Questions this raises.
How would I know a paper was retracted?
Open it at the publisher's site, where a notice appears on the record. Databases vary in how quickly they propagate that, and a PDF saved to your own machine will usually show nothing at all. This is the specific reason a downloaded copy is not sufficient for a source your argument depends on.
Can I ever cite a retracted paper?
Deliberately and explicitly, where the retraction itself is what you are discussing, or where you are documenting how a claim spread. Label it as retracted in the sentence rather than only in the reference. What is not acceptable is citing it as evidence for the finding it originally reported.
What is an expression of concern?
A notice that a journal is investigating a paper without having reached a conclusion. Treat it as a reason to find a different source rather than as a caveat to mention, because the investigation may end in retraction after your work has been submitted and you will have no way to revisit it.
Do retractions happen in every field?
They do, at different rates and for different reasons. Health sciences see more of them partly because more eyes are on the literature and partly because the stakes prompt scrutiny. The checking habit is worth having regardless of discipline, because the cost of the error does not vary.
Is there a fast way to check a whole list?
Open each source at the publisher as you add it to your library rather than checking everything at the end. Built into collection, it costs seconds per source. Done as a separate pass over forty references the night before submission, it is an hour you will not want to spend.
Dr. Sigrid Nyholm
PhD, Nursing, on the evidence-based practice side of the hall. Writes for The Footnote and coaches the students who bring this work to the desk. The rest of the specialists.
Elsewhere in the hall.
Send the assignment and the deadline.
A specialist credentialed in your own field reads what you actually have and replies within two hours with a plan and a price. The reading costs nothing, and if the honest answer is that we are not the right hall for it, you will get that instead.
