What Is the Predictive Ability and Academic Impact of the Altmetrics Score and Social Media Attention?

Background: Citation rate and journal impact factor have traditionally been used to assess research impact; however, these may fail to represent impact beyond the sphere of academics. Given that social media is now used to disseminate research, alternative web-based metrics (altmetrics) were recently developed to better understand research impact on social media. However, the relationship between altmetrics and traditional bibliometrics in orthopaedic literature is poorly understood. Purpose: To (1) assess the extent that altmetrics correlate with traditional bibliometrics and (2) identify publication characteristics that predict greater altmetrics scores. Study Design: Cross-sectional study. Methods: Articles published in The American Journal of Sports Medicine (AJSM), The Journal of Bone and Joint Surgery, Clinical Orthopaedics and Related Research, Acta Orthopaedica, and Knee Surgery, Sports Traumatology, Arthroscopy between January 2016 and December 2016 were analyzed. Among the extracted publication characteristics were journal, number of authors, geographic region of origin, highest degree of first author, study subject and design, sample size, conflicts of interest, and level of evidence; number of references, institutions, citations, tweets, Facebook mentions, and news mentions; and Altmetric Attention Score (AAS). Multivariate regressions were used to determine (1) publication characteristics predictive of AAS and social media attention (mentions on Twitter, Facebook, and the news) and (2) the relationship between AAS and citation rate. Results: A total of 496 published articles were included, with a mean AAS of 8.6 (SD, 31.7; range, 0-501) and a mean citation rate of 15.0 (SD, 16.1; range, 0-178). Articles in AJSM (beta = 19.9; P < .001), publications from North America (beta = 8.5; P = .033), and studies concerning measure validation/reliability (beta = 25.5; P = .004) were independently associated with higher AAS. Greater AAS score significantly predicted a greater citation rate (beta = 0.16; P < .0001). The citation rate was an independent predictor of greater social media attention on Twitter, Facebook, and the news (odds ratio range, 1.02-1.03; P < .05 all). Conclusion: AAS had a significant positive association with citation rates of articles in 5 high-impact orthopaedic journals. Articles in AJSM, studies concerning measure validation and reliability, and publications from North America were positively associated with greater AAS. A greater number of citations was consistently associated with publication attention received on social media platforms.