An Analysis of the 50 Most-Cited "Uveitis" Articles Published Between 2010-2020 From a Bibliographic and Altmetric Perspective

Purpose We aimed to evaluate the 50 most-cited articles on uveitis according to their Altmetric Attention Score (AAS) and additional metrics. Methods The Web of Science (WoS) core collection database was used in this study. The article and journal metrics and characteristics were examined. In addition, the effect of article and journal metrics on the AAS was examined with multivariate adaptive regression splines (MARS). Results The number of citations of the evaluated articles ranged from 670 to 90, and AASs ranged from 633 to 0. According to the MARS model, the importance scores of the predictors were as follows: article influence score (100%), immediacy index (77.74%), number of years since publication (57.79%), times cited in WoS (32.69%). We found that the trend of articles on uveitis was related to the "treatment category", namely, adalimumab. Second, the popular topic was uveitis caused by viruses. Conclusions We found that citation-based metrics and year of publication contributed to AAS. AAS appears to be inadequate in assessing the quality of articles. However, due to the electronic transformation of the publishing industry, it seems inevitable that altmetrics will become an additional supportive metric.


Introduction
Uveitis refers to inflammation of the uveal tract, which can cause loss of vision [1]. The incidence of uveitis peaks at working ages [2]. Blindness caused by uveitis is treatable or potentially preventable. Not all types of uveitis have the same effect on vision, and uveitis should be classified and treated according to the underlying etiology. Systemic immunomodulatory agents and conventional immunosuppressive agents are available for the treatment of non-infectious uveitis (NIU). In recent years, the introduction of biological agents has led to important developments in the treatment of uveitis. Biological agents are promising drugs designed to be effective based on the molecular understanding of disease pathogenesis [3]. Infectious uveitis (IU) is one of the most common and visually devastating causes of uveitis worldwide [4]. IUs can mimic NIUs and their early diagnosis and treatment are important.
The main purpose of authors is to convey scientific information to a wide audience. The impact of the articles is evaluated by various metric measurements. The most important of these is the number of times cited. Another important point is the metrics of the journal publishing the article. The most well-known of these is the impact factor (IF), which is calculated by examining the citations of the articles in scientific journals and showing the impact of the journals in their fields [5].
Citations are of course important in calculating journal metrics. However, nowadays, due to the development of the internet and social media, scientific articles are not only cited from scientific journals but also from many sources. Considering that it would be incomplete to evaluate the popularity of a scientific publication by the number of scientific citations alone, altmetrics were created that include citations in channels such as social media, Twitter, blog posts, and news sites. Altmetrics compile the impact of the article on the internet and social media from dozens of different sources and scores its popularity. Altmetrics also measure where and by whom the article is cited, such as language, country, and region [6]. 1 1 1 In this study, we aimed to evaluate the 50 most-cited articles on uveitis according to their Altmetric Attention Score (AAS) and additional metrics. The journals in which the articles were published, categories of the journals, the types of articles, and the topic of the article were examined. In addition, we aimed to show the effect of article and journal metrics on the AAS.

Materials And Methods
The Web of Science (WoS) core collection database (www.webofknowledge.com) was used in this study. In the advanced search section, the term "Uveitis" was searched in the title, abstract, and keywords. The 50 most-cited articles were included in the study (Access date: February 17, 2022). The time filter was applied between January 01, 2010, and December 31, 2020. No other filters were applied. All data obtained were collected on the date of access. The list of articles obtained is presented in the appendices. Non-relevant articles and animal studies were excluded. Relevant metrics of all articles were obtained. We focused on articles whose primary priority was uveitis.
Titles, first authors, all authors, year of publication, number of years since publication (NYsP), number of citations, Altmetric Attention Score (AAS) and average citations per year (ACpY), journal IF, journal 5-year IF, Q category, journal citation indicator (JCI), Eigenfactor score (EF), article influence score (AIS), immediacy index (II) and h-index were recorded and examined.
The AASs of the articles were obtained from the bookmark "Altmetric it!" (www.altmetric.com). The h-index of journals was obtained from scientific journal rankings (www.scimagojr.com). All other metrics were obtained from the WoS and journal citation report 2020 (www.jcr.clarivate.com

Results
The number of citations of the evaluated articles ranged from 670 to 90, and AASs ranged from 633 to 0. Detailed metrics and characteristics of the first 50 uveitis articles are summarized in Table 1. The top 50 articles were cited 7897 times. The relationship between the year of publication of the articles and the annual cumulative citation is presented in Figure 1. The scatter plot of AAS by years is presented in Figure 2.   The journal with the most articles in the top 50 was "Ophthalmology", with 10 articles. The list of journals in which the top 50 uveitis articles were published and their detailed metrics are presented in Table 5 in the appendices. Journals in the ophthalmology category were the most published category with 29 articles and the median citation of WoS and median AAS were 126 (104.5-161.25) and 2 (0.25-7.75), respectively. Journals in the general medicine category had the highest median citations of 255.5 (154.0-279.25) and an AAS median of 209 (29.0-393.0). Elsevier was the publisher with the most articles, with 19 articles. The metrics by category of the journals in which the first 50 uveitis articles were published are summarized in Table 2.

TABLE 2: Journal metrics and changes by Q category of WoS and metrics by journal category
Median (25%-75% interquartile range), NA: not applicable AAS was moderately correlated with IF, EF, AIS, JCI, and ACpY. AAS was weakly correlated with the number of citations and II. The correlation between AAS and article and journal metrics is summarized in Table 3. In the MARS model, the predictors are listed according to their importance scores calculated on the 100% scale, and the most important variable always receives a 100% score. Accordingly, the importance scores of the predictors are as follows: AIS (100%), II (77.74%), NYsP (57.79%), times cited in WoS (32.69%). In the model constructed with predictors, R2 was 0.97. The contribution of the interaction of the predictors to the model is presented in Figure 3.

Discussion
In this study, in which we evaluated articles related to uveitis, we found that the AAS was associated with citations, citation-based metrics, and the number of years since publication. The fact that AIS were the most important predictors indicates the importance of publishing an article in top journals. The fact that the II is the second-most important predictor reveals that the articles in the journals, which attract the attention of the scientific world faster, are more circulated in the online spaces. This was demonstrated by the fact that NYsP was the third-most important predictor in the regression model. However, caution should be exercised when interpreting these results. There was a time filter in the search strategy in this study. The NYsP might have been a more important predictor without the time filter since the provision of altmetric data began in 2011 [7]. Another thing to be aware of is the existence of social media accounts where journals share articles with their followers. In this way, they aim to make it possible for an article to reach a wider audience and attract attention [6,8]. Finally, there may be predictors (e.g. popular topics) not included in the regression model.
The topic of most public interest may differ from that of the scientific community, and a more-cited article may not receive enough public attention [9]. Eight articles did not have high AAS in the altmetric analysis of glaucoma [10]. The fact that eight articles in the present study did not have high AAS is concrete proof that the agenda of the public and the scientific community is different. On the other hand, in the present study, it was observed that two case reports of uveitis caused by Zika virus and Ebola virus had a high AAS. The fact that case reports with a low level of scientific evidence compared to others have such a high AAS can be considered a result of the popularity of a disease affecting the public. In the study in which retinal articles in ophthalmology journals were evaluated, it was found that an article on retinal complications of Zika virus had the highest AAS [11]. In a study in which an altmetric analysis of 100 articles related to COVID-19 was performed, it was reported that the AAS was 3246 ± 3795 (85-16548) [12]. In a study evaluating the altmetric activity of 12.3 million WoS publications, it was reported that infectious diseases with a social impact received more attention [13]. AAS can also be interpreted as a social IF [14], and when this information is evaluated, we can say that topics that create social impact get more public attention than topics that are more technical. We can deduce that AAS may be low for a specific topic that concerns healthcare professionals. However, it should be kept in mind that as the popularity of the topic decreases, the speed of altmetric activity may slow down.
When the 10-year pool of articles included in the study is evaluated, it is seen that the majority of the articles are composed of NIUs, and treatment topics are predominant. In the past decade, the introduction of biological agents in the treatment of NIU has revolutionized the treatment. We noticed a concentration of treatment-related articles on adalimumab. Patients with uveitis are a relatively younger population, and uveitis has a more widespread effect on ocular morbidity than age-related macular degeneration [15]. Adalimumab is presented as a promising new treatment option with improvements in visual functionality outcomes [1,16]. The article with the second-highest AAS (313) in the present study is associated with adalimumab. It may be due to patients wanting to know and understand their treatment, as well as wanting to explore new treatment options. On the other hand, journals in the general medicine category had the highest AAS. The fact that these non-ophthalmology journals are on our list and have a high AAS is proof that they are effective in both the ophthalmology community and the public.
The first limitation of our study was that we did not use a large data sample in the analysis. The second limitation was that we used a time filter. The third limitation was we used single-search terms. Despite these limitations, we have demonstrated the importance level of the effect of additional metrics on ASS.

Conclusions
In conclusion, first, we found that the trend of articles on uveitis was related to the "treatment category"adalimumab. Second, we found that the most popular topic was uveitis caused by viruses. Finally, we found that citation-based metrics and year of publication contributed to AAS. AAS appears to be inadequate in assessing the quality of articles. However, due to the electronic transformation of the publishing industry, it seems inevitable that altmetrics will become an additional supportive metric.
Appendices  show the list of articles and journals included in this paper, respectively.

Additional Information Disclosures
Human subjects: All authors have confirmed that this study did not involve human participants or tissue. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.