Using Big Data to Assess Legitimacy of Plastic Surgery Information on Social Media
Abstract Background The proliferation of social media in Plastic Surgery has posed significant difficulties for the public in determining legitimacy of information. In this work, we propose a system based on social network analysis (SNA) to assess the legitimacy of contributors of information within a Plastic Surgery community using academic Plastic Surgery and one social media outlet as a model. Objectives The aim of this study was to quantify the centrality of individual or group accounts on Plastic Surgery social media. Methods To develop the model, we chose one high-fidelity, active, and legitimate source account in academic Plastic Surgery (@psrc1955, the Plastic Surgery Research Council) on one social media outlet (Instagram, Facebook, Menlo Park, CA, USA). We then recorded all follower-following relationships between accounts and used Gephi (https://gephi.org/) to compute five different centrality metrics for each contributor within the network. Results We identified 64,737 unique users and 116,439 unique follower-followed relationships within the academic Plastic Surgery community. Among the metrics assessed, the in-degree centrality metric is the gold standard for SNA, hence we designated this metric as the Centrality Factor (CF). Stratification of 1000 accounts by CF demonstrated that all of the top 40 accounts were affiliated with a Plastic Surgery residency program, a board-certified academic plastic surgeon, a professional society, or a peer-reviewed journal. None of the accounts in the top decile belonged to a non-plastic surgeon or non-physician, however, this increased significantly beyond the 50 th percentile. Conclusions This study took a data-driven approach to identifying and vetting a core group of interconnected accounts within one Plastic Surgery sub-community for the purposes of determining legitimate sources of information.