Post-publication Activity on PubMed Commons: Evaluation of Comments from 3 Months in 2014

Sunday, February 14, 2016
Melissa Vaught, Contracted to the National Center for Biotechnology Information, National Library of Medicine, National Institutes of Health, Bethesda, MD
Background: PubMed includes more than 25 million citations for biomedical literature, with 3-4 million users per day. In 2013, the National Center for Biotechnology Information (NCBI) – part of the National Library of Medicine (NLM) at NIH – launched PubMed Commons as a pilot initiative to enable signed commenting on publications in PubMed. As of June 2015, more than 9000 members had joined PubMed Commons, with 10% posting comments. Members had contributed 3319 comments on more than 2800 publications. A representative subset of comments was selected to study how members were using PubMed Commons. Methods: A detailed evaluation was conducted for active comments posted to PubMed Commons from July 1 to September 30, 2014 (414 comments on 365 publications). Information was collected on publications to which comments were posted, including publication date, funding sources, general topic, and type of article. Two reviewers assessed the primary and linked content of comments independently and subsequently discussed assignments to resolve conflicts and consolidate categories. Author response rates were also calculated. Results: Just over 60% of publications with comments had been published in PubMed in the previous year. Two-thirds of comments were posted to primary research articles. About half of the comments were posted to medical- and societal- health-related publications, with clinical trial reports representing 13% of publications with comments. More than half of the comments posted represented discussion or critique of publications by commenters who had not contributed to the publication. Authors responded to only 8% of comments posted. Authors more commonly initiated comments on their own publications to provide access to datasets, a version of the article, or further related information. Conclusions: The use of PubMed Commons is low but makes valuable contributions to post-publication curation and discussion.