Identifying Retracted Journal Articles

A continuing hot topic in the news is the increasing number of scientific studies that have been discovered to be wrong and are retracted by their publishers. Indeed, Nature News reported that more than 10,000 research papers were retracted in 2023. Citing retracted articles should be avoided in scholarly publishing as it can decrease the credibility and reliability of your work.

Why is this important? Articles can be retracted for a number of reasons but typically because of either error (intentional or unintentional) or fraud (Brainard and You, 2018); however, they often continue to be cited because authors are not aware the article was retracted. Journals are increasingly reporting retractions and taking them seriously. But how does the author know when an article has been retracted?

Recently, the Medical Center Library was asked by a Duke researcher about an email from RetractoBot with the subject line: “You cited a retracted paper in your Hypertension paper published in 2020.” The researcher was wondering if we had seen this before and if anything needed to be done? It turns out the researcher was sent the email as part of a randomized control trial that aims to make retraction notices more visible by authors so that retracted papers are no longer mistakenly cited.

What is RetractoBot? RetractoBot automatically emails authors when articles they have cited are retracted. You can find out more in their study protocol.

What other retraction alert tools and services exist?  
In 2010, the Retraction Watch website was created to track retracted articles and to spotlight how many scientific articles are withdrawn each year and why. In 2018, they launched the Retraction Watch Database which was acquired by Crossref in 2023 and is now freely available. This database currently contains over 47,000 retracted articles tagged with reason(s) for retraction and is the underlying source of data for numerous scholarly studies, and new tools used to alert authors if publications they have used have been retracted. For example, citation managers such as EndNote and Zotero have integrated the Retraction Watch database so that users are alerted if any of the publications in their library are retracted.  

What do I do when an article I have cited has been retracted? So what do you do if you find out either from RetractoBot, or an alert in EndNote or Zotero, that an article you have cited in a publication has been retracted? If the retracted article was cited as part of your introduction or background, it may or may not impact the conclusions of your study. However, if the cited article was an included study in your review or systematic review or meta-analysis, then it certainly would impact the conclusions of your study. Retraction Watch has some good advice for authors, which includes reaching out to the journal editor to see their recommended course of action. For more advice on what to do, see this article by Retraction Watch.

In the meantime, rely on your citation software to alert you to whether any articles in your library have been retracted. It is also a good idea to check your references (like in the Retraction Watch database) before submitting a manuscript. If your article is specifically discussing a specific retracted paper, then best practice is to cite both the original article and the retraction notice which should have different DOIs.

If you have questions, please reach out to the Medical Center Library & Archives at medical-librarian@duke.edu