Altmetric Activity and Scientific Integrity: A Retraction Analysis of PubMed-Indexed Social Media-Related Publications

Résumé fourni par la source

Background: Social media is an essential tool for the dissemination of scientific information.The growing number of publications related to social media raises concerns about the proliferation of misleading information in the scientific literature.Retractions act as a crucial corrective mechanism in scholarly publishing; however, their relationship with social media visibility has not been thoroughly examined.Methods: As part of our descriptive research, 55 retracted social media-related publications indexed in the PubMed database were analyzed.Such characteristics as the number of authors, the time from the date of publication to the retraction, the indexing of publications in the Web of Science, Scopus, and Directory of Open Access Journals databases, the country of the authors of correspondence, the reasons for the retraction, and Altmetric Attention Score (AAS) were analyzed.The data on citations and mentions in various social media platforms were analyzed.Results: The highest number of retracted publications was published in 2023 (n = 39), and the median time from publication date to retraction amounted to 541 (28-912) days.China and India were the leading countries for corresponding authorship and international collaborations.The leading causes of retractions were Investigation by Journal/Publisher (n = 45), Unreliable Results and/or Conclusions (n = 41), Concerns/Issues about Referencing/ Attributions (n = 37), Concerns/Issues with Peer Review (n = 35), and Concerns/Issues about Results and/or Conclusions (n = 35).The median AAS of retracted publications was 1.5 (1-1,627).A positive correlation was found between the number of citations and AAS (rho = 0.562, P < 0.001).Conclusion: Retracted social media-related papers can gain substantial digital circulation before formal retraction, highlighting vulnerabilities in the current scientific interaction ecosystem.These findings urge more post-publication supervision, timely retraction notices, and international collaboration to enhance research integrity, especially in the rapid, decentralized digital distribution age.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.