Presence of social media mentions for vascular surgery publications is associated with an increased number of literature citations

Abstract

Objective: The past two decades have seen a vast expansion of social media in all aspects of our lives. Scholars and journals are steadily increasing their social media presence to reach a wider audience. We compared the social media mentions (SMs) of vascular surgery publications and their effect on the literature citations (LCs) for them.

Methods: A total of 169 articles from three renowned vascular surgery journals (Journal of Vascular Surgery [JVS], Annals of Vascular Surgery, and European Journal of Vascular and Endovascular Surgery) in October 2016 were collected. All three journals are published by the same publisher (Elsevier). SMs were tracked using Altmetric Bookmarklet for Twitter and Facebook mentions. The LCs were evaluated using Scopus and Google Scholar. The number of citations was compared between those with and without any SMs and among the three journals using nonparametric Kruskal-Wallis tests. The proportion of articles with SMs was compared among the three journals using a χ² test. The relationship between the numbers of SMs and LCs was assessed using the Spearman rank correlation coefficient and reported as 95% confidence intervals. Statistical significance was assigned at P < .05.

Results: Of the 169 articles examined, 51 (30.2%) had a presence regarding social media usage. JVS has both Twitter and Facebook presence. The Annals of Vascular Surgery and European Journal of Vascular and Endovascular Surgery only have Twitter accounts. JVS had the highest total number of citations, number of LCs per manuscript, and SMs per manuscript. A significant difference was found in the median, Q1 (median of the lower half of the data), and Q3 (median of the upper half of the data) number of total Google citations between those articles with and without SMs (median, 8.0; Q1, 3.0; Q3, 17.0; vs median, 3.0; Q1, 0.0; Q3, 8.0, respectively; Kruskal-Wallis P < .001). Similarly, a significant difference was found in the median number of total Scopus citations between those articles with and without SMs (median, 5.0; Q1, 2.0; Q3, 13.0 vs median, 2.0; Q1, 0.0; Q3, 6.0, respectively; Kruskal-Wallis P < .001). Articles with a SM showed a 2.7- fold increase in median total citations in Google and a 2.5-fold increase in median total citations in Scopus. The Spearman correlation coefficients to determine the relationship between the absolute number of SMs and LCs revealed a positive, but weak, correlation, largely driven by the majority of articles with no SMs. The difference in the median number of citations among the three journals was not statistically significant, either by Google (P = .22) or Scopus (P = .08), nor was the difference in the proportion of articles with SMs among the journals statistically significant (P = .36).

Conclusions: The presence of SMs for vascular surgery publications, especially clinical science articles, was associated with a significantly increased number of median LCs during the 3 years after publication. The three journals did not differ with respect to the median number of citations or proportion of articles with SMs.

Keywords: Citations; Impact factor; Social media; Vascular surgery.