Tea Marasović,1 Ana Utrobičić,2,3,4 Ana Marušić3,4
Recently updated International Committee of Medical Journal Editors (ICMJE) recommendations suggest correcting honest errors by “retraction with republication of the changed paper, with an explanation.” MEDLINE uses “Corrected and Republished Article” to indicate correction of “a previously published article by republishing the article in its entirety.” We assessed how other bibliographical databases indexed article corrections with replacement.
Articles indexed as “Corrected and Republished Articles” in MEDLINE from January 2015 to December 2016 (n = 29) were analyzed for the information presented in journals and in the Web of Science (WoS) and Scopus. Two authors independently extracted the data and reached a consensus for disagreements.
Twenty-nine articles were published in 24 biomedical journals from different research areas (all articles were indexed in Scopus and all but 3 indexed in WoS; median Impact Factor for WoS journals, 2.98; 95% CI, 2.15-3.35). Half of the journals published a separate item to indicate correction, and half had a CrossMark tag on the corrected article (Table); CrossMark tags had no links to the original article or notification of correction. PubMed did not provide the links between the corrected to the original article in 1 case. Web of Science and Scopus indexed corrected articles most often as a correction (WoS) or erratum (Scopus). Five articles in WoS (17%) and 11 in Scopus (38%) were indexed as “articles” in the same way as original articles, which made it difficult to differentiate between the versions. When corrected articles were indexed, they often lacked links to the original articles. Original and corrected articles had a similar median number of citations (WoS: original, 2.0; 95% CI, 2.0-4.2; corrected, 2.0; 95% CI, 1.0-6.8; Scopus: original, 2.0; 95% CI, 1.0-4.4; corrected, 2.0; 95% CI, 1.0-10.2). A PubMed search for “retraction and replacement” identified 5 more articles published in JAMA, which were indexed only as retracted and not as corrected and republished publications. Web of Science indexed only 1 version of these articles, and the notices of retraction and replacement were indexed as a “letter,” “correction,” or “editorial material.” Scopus also indexed all 5 articles, and notices were indexed as a “letter” or “erratum” (1 was missing).
There seem to be serious discrepancies in indexing corrected and republished articles in major databases, which diminishes the credibility and transparency of the research and publication system. While scientific self-correction should be supported, all stakeholders in the publication process should commit to ensuring that published scientific articles are appropriately indexed, interlinked, updated, and/or amended in a timely and efficient manner.
1Research Office, University of Split School of Medicine, Split, Croatia; 2Central Medical Library, University of Split School of Medicine, Split, Croatia; 3Cochrane Croatia, Split, Croatia, email@example.com; 4Department of Research in Biomedicine and Health, University of Split School of Medicine, Split, Croatia
Conflict of Interest Disclosures:
Dr Marušić is a member of the Peer Review Congress Advisory Board but was not involved in the review or decision for this abstract.
This study was funded by grant IP-2014-09-7672 from the Croatian Science Foundation.
Role of the Funder/Sponsor:
The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the abstract.