Discussion
Once the analysis of the online attention of the 23,179 original articles detected in WoS was performed, it is important to consider the definition of TE by Langer and Vacanti in 1993 (Langer & Vacanti, 1993). They defined TE as an interdisciplinary field that applies the principles of engineering and life sciences toward the development of biological substitutes that restore, maintain or improve tissue function. The onset and progress of TE have led to a revolution in health science practice that has also supposed a shift in contemporary medical paradigm. This revolution requires an objective quantification, becoming the employ of bibliometrics a useful tool (Grant, Cottrell, Cluzeau, & Fawcett, 2000).
In this way, we have performed an altmetric-based analysis of the core documents retrieved from WoS. First, the descriptive analysis showed the distribution of TE original articles within seven web-based platforms (WoS, Altmetric Attention Score, Twitter, Number of Mendeley readers, Facebook, Patents and News). Concerning the results reported above, the evolution of TE diffusion in Twitter stands out within the so-called social networks becoming crescent from 2012 to nowadays. The reasons for this growing probably lie in two elements: a larger diffusion of knowledge and a higher academic impact.
Regarding this larger diffusion of knowledge, the structure of Twitter, a micro-blogging platform that enables the user to communicate short messages with their virtual colleagues, has developed a singular model of scientific communication with an especial information flow (Finin, Tseng, Akshay, & Xiaodan, 2007; Kwak, Lee, Park, & Moon, 2010). A study conducted by Kwak et al. demonstrated that the retweet constitutes the nucleus of this new model of communication and, thus, retweets on TE documents can spread the information beyond the limits of their original authors, expanding them to the followers’ networks (Darling, Shiffman, Cote, & Drew, 2013; Huberman, Romero, & Wu, 2009).
One of the consequences of this new model of communication is the spreading of new medical approaches for the treatment of severe untreated diseases. The information of possible new treatment, as happens in TE products, may reach through the patients to primary care physicians. For this reason, this situation is being considered in the future training of family medicine residents (Sola et al., 2019).
Respect to the consequently higher academic impact, a study conducted by Eysenbach showed that highly tweeted articles are 11 times more likely to end up being highly cited and, thus, that Twitter correlates with traditional metrics of scientific impact (Eysenbach, 2011). The upward trend of TE original articles on Twitter is probably also related to a higher academic impact of those TE mentioned documents.
Moreover, relevant information can be extracted when comparing TE original articles ranked by WoS citations and Altmetric Attention Score. Obtained results demonstrate that TE academic and social interest do not follow the same path. These results, firstly demonstrated for TE discipline, are similar to other research fields and evidence a profound discrepancy between the academic focus of interest and social assumption of scientific advances (Choo et al., 2015; Gunn, 2013) . Then, the full attention of TE research may not be well addressed through traditional metrics. According to Bornmann, citations only assess the impact of scholarly literature on those who cite, and this neglects many audiences of scholarly literature who may read, but do not cite as “pure” readers (Bornmann, 2015).
The correlation results show that citations of TE original articles in WoS are well correlated to the Number of Mendeley readers. It can be explained partly attending to the own nature of Mendeley, based on a community of bibliographic users. As a platform designed to store and share references, the use of Mendeley has been previously correlated to future citation counts in other biomedical sciences (M. a. W. Thelwall, P., 2016). In this way, this correlation also occurs in TE research area. Nevertheless, Mendeley users do not have to be publishing academics, but may also be practitioners or students (Haustein, Larivière, Thelwall, Amyot, & Peters, 2014; Mohammadi, 2014). Therefore, the correlation could be related to a broader spectrum of TE scientific activity not only restricted to the academy.
Positive but weaker correlations were obtained for platforms such as Twitter, News and Blogs, accounting for a more accessible and open to the public kind of scientific impact. The appearance of TE original articles in Video, Reddit and Peer Review implies a fewer WoS citation count for TE original articles given that negative correlation data were observed. Probably, it could be influenced by the nature of this kind of platforms. For example, Reddit is a platform in which the virality is a crucial factor (Haralabopoulos, Anagnostopoulos, & Zeadally, 2015). According to Berger and Milkman (Berger & Milkman, 2012), those contents that evoke emotions of activation (e.g. anger, awe, anxiety) are more suitable to become viral, in contrast to deactivating emotions (e.g. softness). Therefore, papers could be mentioned to be criticized or are reporting findings that are surprising or shocking, but with low interest for the academic community.
Nevertheless, according to Thewall, correlation results could obscure relationships between variables, especially if there is one strong one (Mike Thelwall & Nevill, 2018); consequently, an exploratory factorial analysis was performed. Six factors were identified in the factorial analysis. From the analysis of these factors, two clear groups can be extracted: Academic nature (Factor 1) and social nature (Factor 4-Factor 6). Each factor accounts for an aspect of TE online attention; i.e. Factor 1 is related to traditional scholarly impact as WoS citations and Number of Mendeley readers joint together; Factor 4-Factor 6 gather different platforms that covered the social diffusion of the science. The relationship of the different platforms in F1 is not surprising since the nature of the users is similar; they are “the spot” of researchers. There are other factors which relationship is not clear. Therefore, we cannot explain the possible influence that has in the final academic impact. Factor 4-Factor 6 are the platforms of interest for the present study, they comprise some media in which the author can present their findings and try to reach the population.
Moreover, News and Facebook appear together suggesting the evidence of a newsworthiness factor while Twitter constitutes a separate one. A possible interpretation is that Twitter has a leading role in TE online attention: historical and cognitive reasons can be argued. On the one hand, the development of TE during XX century has taken place in parallel with the burst of social media and probably TE researches have substituted the idea of the academic community for the virtual department (Pogorielov, 2017; Xuemei Li, 2012). In the other hand, the structural multidisciplinary of TE (10) can be ideally appropriately displayed using social networks such as Twitter and the relations between industry, academics and clinicians could be improved without temporal or geographical restrictions (Bik & Goldstein, 2013; Kwak et al., 2010).
Lastly, Blogs are a tie in a unique factor with Peer Review mentions. In TE, blogs constitute an active space for knowledge exchange (Brown & Woolston, 2018) and to communicate science to major stakeholders (Weigold, 2001) . The association within a common factor of Peer Review and Blogs could be stated for a major reason lack of elucidation. Nevertheless, according to Weigold, the sharing of well-constructed information online contributes to informing society about real possibilities of scientific progress (Weigold, 2001); and it constitutes a pillar in TE because it offers a new therapeutically scenario for the treatment of several diseases (Li et al., 2016).
Finally, we aimed to develop a model to discover the different online influences that determine the future TE citation counts. As Thelwall declared, it is reasonable to consider Altmetric.com scores in conjunction with journal impact to get an idea of which articles are more likely to attract longer-term citations (Mike Thelwall & Nevill, 2018). Accordingly, we obtained a regression equation to derive 2018 citation counts from 2015 Almetrics.com scores for TE original articles from 2015 that Altmetric tracked. The predictive power (R2) of the model was 10.7% when Mendeley readers were not added as an independent variable. When Mendeley readers were considered, a value of 41.4% for R2 was achieved, increasing the model accuracy. All regressions were statistically significant. Consequently, 2015 altmetric scores for TE original articles account for almost half of the variability in future citation counts. It follows that altmetric scores are useful if TE researchers aim to discover future citation counts. Furthermore, the different actors involved in the TE scientific diffusion should consider implementing strategies to be present in the different platforms that increase the final scientific impact.
Although the findings provided in the present paper are interesting, several limitations have to be addressed. First, only a percentage of the publications indexed in WoS is available in Altmetric.com. Therefore, the conclusions are influenced by the core obtained. Second, the factorial analysis is performed only in one year; although the behavior of the research area could be similar, it could be influenced by the published topics or other factors. Finally, the intentional tweeting by the publisher or the editor of the journal was not analyzed.