Published Research on COVID-19 in the Eastern Mediterranean Region: Bibliometric Analysis

Background: The challenges presented by the COVID-19 pandemic have led to unprecedented global research activity. The Eastern Mediterranean Region (EMR) continues to contribute to COVID-19 research driven by the unique challenges of the region, including the protracted conflicts, already stressed health systems, and serious health and social inequalities. Objective: This study aims to provide an overview of the publication activities and trends in COVID-19 research in the EMR from the onset of the disease to early 2022 using bibliometric methods. Methods: A literature search using Scopus was conducted from December 1, 2019, to January 31, 2022, using keywords relevant to COVID-19 and the World Health Organization (WHO) EMR country list. Data were exported and analyzed using Microsoft Excel and the Citation Overview function on Scopus. The quality of journals was determined using SCImago Journal Rank and CiteScore. VOSviewer software was used to visualize the relationships between authors, countries, and key terms used in the retrieved documents. Results: A total of 6880 documents were retrieved, of which 1805 (26.24%) were from the Kingdom of Saudi Arabia (KSA) and 1782 (25.90%)


Introduction
The COVID-19 pandemic continues to be a major global challenge, placing a heavy burden on the economy, social life, public health systems, and the delivery of health services in all countries worldwide [1]. Despite the rapid development of various vaccines to prevent severe illness and treatment modalities, the world still faces numerous challenges in controlling COVID-19 and its impacts [1]. Disparities in global access to vaccines, increasing vaccine hesitancy, impractical long-term implementation of preventive public health measures, and the continuous emergence of new variants are some of these challenges [1][2][3]. The Eastern Mediterranean Region (EMR) remains at high risk of COVID-19 case surges with the associated short-and long-term consequences primarily due to the combination of the aforementioned causes, added to which are the specific regional environment with protracted conflicts; high denial attitude toward the pandemic; chronically stressed health systems; and the presence of health, economic, and social inequities [4][5][6][7].
As of April 1, 2022, there have been 21,576,432 confirmed cases and 340,628 deaths attributable to COVID-19 in the EMR [8]. The EMR has a population of nearly 700 million people in 22 countries with high susceptibility to infectious diseases due to numerous political, economic, social, cultural, and human-animal interactions [5,9]. Almost 40% of the world's population in need of humanitarian assistance lives in the EMR and is at high risk of coronavirus transmission and associated severe consequences due to overpopulation, suboptimal sanitation, a high caseload of noncommunicable diseases (NCDs), and limited resources and health system capacity [5]. Furthermore, a number of EMR countries have limited testing facilities, weak health system infrastructure and response, and inadequate vital registration and documentation, all of which have been associated with the possibility of underreporting or undertesting or both of COVID-19 cases [4,5,10]. The introduction of vaccination programs in the EMR continues to be hampered by logistical, economic, security, and population hesitancy issues, as well as reaching people living in hard-to-reach areas [11].
COVID-19 had a strong impact on scientific publications worldwide, as evidenced by the rapid increase in the volume and pace of COVID-19 publications, with many journals dedicating special sections or issues to COVID-19, even at the expense of other topics [12,13]. The COVID-19 crisis has also stimulated an unprecedented level of multidisciplinary research involving not only those with a direct interest in COVID- 19 research and health, such as virologists, epidemiologists, and clinicians, but also researchers from diverse fields, such as artificial intelligence and business [14,15]. The projected long-term impact of COVID-19 and the fact that it affects multiple areas of life have led to a burst of studies examining the impact of COVID-19 on many nonmedical and nonclinical research areas, such as education, business and management, tourism, and agricultural food supply [16][17][18][19]. The global research funding mechanisms and magnitude have also changed. Between the beginning of January 2020 and July 2021, more than US $21.7 trillion was reportedly committed globally to different COVID-19 activities according to data analysis on Devex's funding platform [20]. Some governments with limited fiscal space or limited prearranged funding sources had to reallocate their existing budgets and allocate funds to COVID-19 studies, even at the expense of other disease control programs, such as for malaria and HIV/AIDS in Africa [21,22]. In the EMR, COVID-19 research has flourished, with countries and organizations dedicating special effort, in addition to technical and financial support for such research [23,24]. Our aim is to provide an overview of the published research activities and trends in COVID-19 research in the EMR by applying bibliometric methods to identify research collaborations, trends, and emerging research themes.

Search Strategy
We conducted a literature search on January 31, 2022, using the Scopus database, covering the period from December 1, 2019, to January 31, 2022. Scopus was chosen for its data mining and bibliometric functions, as well as its compatibility with Visualization of Similarities Viewer (VOSviewer) software. Additionally, Scopus is the largest indexing database, combining the characteristics of both PubMed and Web of Science, and thus allows for enhanced utility, both for literature research and academic needs, including citation analysis [25]. As Scopus does not use subject headings but instead assigns index terms to papers, our search query included the field code KEY, which combines both author keywords and indexed terms [26]. Our search query included keywords relevant to COVID-19 and the World Health Organization (WHO) list of countries in the EMR (Multimedia Appendix 1) [27]. Keywords for COVID-19 included "severe acute respiratory syndrome coronavirus-2," "SARS-CoV-2," "sars-2," and "2019-nCoV," as well as the words "novel," "new," "2019," "Wuhan," "Hubei," and "China" adjacent to the words "coronavirus" and "COVID." We additionally used wildcards and truncations, as needed, to refine the search. We limited the search to papers, reviews, chapters, and books published in English, associated with any of the EMR countries. We included both published papers and papers in press (accepted for publication by a journal but not assigned to a specific journal issue).

Citations and Quality Assessment
The Citation Overview function in Scopus enabled us to determine the mean, median, and range of citations of all retrieved documents, as well as identify the topmost cited documents. Additionally, author details were reviewed on Scopus to determine the topmost productive authors' H-index and citation score.
We assessed the quality of the most productive journals on the topic of COVID-19 in the EMR using the 2021 SCImago Journal Rank (SJR), CiteScore (CS), and the Source-Normalized Impact Per Paper (SNIP). The SJR is based on centrality concepts and data from Scopus, and it limits self-citations and falsely inflated quality ranks, while the CS gives a comprehensive, transparent, and current view of a journal's impact [28]. SNIP is also based on data from Scopus, and it measures impact by weighing citations based on the total number of citations in a subject field, enabling direct comparison of sources in different subject fields [29]. Using the SCImago ranking website, we identified the subject area of each journal and its corresponding SJR rank divided into 4 equal quartiles, with Q1 comprising the quarter of the journals with the highest values, Q2 the second-highest values, Q3 the third-highest values, and Q4 the lowest values [30].

Visualization of Similarities
To analyze and visualize relationships among authors, countries, and the key terms used in the retrieved publications, we exported the citation information, bibliographical information, abstracts and keywords, and funding details and included references from Scopus into VOSviewer v.1.6.16 (Centre for Science and Technology Studies, Leiden University) [31].
In constructing the networks and maps, we used a clustering resolution of 1 and a minimum cluster size of 12 to eliminate small clusters, and we illustrated each cluster (eg, group of linked authors, countries, or keywords) using a different color. We scaled the maps based on document weights, unless otherwise specified, so the diameter of each label denotes the number of occurrences of the author, country, or keyword specified by the label, in the documents, and the distance between 2 labels represents the degree to which they are associated [32].
Coauthorship network analysis was performed based on the full counting method. When we used authors as the unit of analysis, we excluded papers authored by >25 authors, and we set both the minimum number of papers published by an author and the minimum number of citations of an author at 5 to identify prominent authors who have published on the topic. When we used countries as the unit of analysis, we excluded papers authored by >25 countries and only included countries with ≥5 published papers. We did not place a restriction on the number of citations during the countries' coauthorship analysis. We identified the number of coauthorship links countries have with one another and the total strength of the links using the "links" and the "total links strength" attributes provided by VOSviewer.
To identify trending topics relating to COVID-19 in the EMR, we used co-occurrence analysis of author keywords occurring at least 10 times in publications. We excluded the names of countries and regions from the list of keywords to focus on scientific themes, and we also excluded all synonyms of COVID-19, which might obscure the results. We applied normalization based on the association strength to eliminate redundancy in similar keywords that define the same concept. Additionally, we mapped the keyword co-occurrence using overlay visualization to determine the evolution of themes with time.

Publication Distribution by Affiliation
Each publication had 1 or more institutional affiliations. Most of the published documents were associated with 117 universities in the EMR, with 396 (5.76%) and 370 (5.4%) of 6880 documents being associated with the Tehran University of Medical Sciences in Iran and King Saud University in KSA, respectively (Table 1) The only international health agency that was associated with COVID-19-related research in the EMR was WHO, being a contributor to 62 (0.90%) of 6880 documents.

Source of Funding
Most of the identified publications did not report a source of funding (5020/6880, 72.97%). However, of those that did, most were funded by King Saud University (282/1860, 15.16%). The Most of the top publishing journals were categorized as medical journals or health-related journals, except for 1 journal, Sustainability Switzerland, which focuses on energy, environmental science, and social sciences.

Citation Metrics
The retrieved documents received a total of 53,516 citations, with an average of 7.78 (SD 34.

Coauthorship
A total of 26,798 authors were identified to have worked on publications with ≤25 authors. Of those, 796 (2.97%) met the threshold of having published ≥5 papers on the topic and being cited ≥5 times. Of the 796, 719 (90.33%) authors were connected in a total of 16 different clusters (Figure 2). Similarly, 285 countries were identified to have coauthored publications (with ≤25 countries per publication). Of those, 82 (28.77%) countries had ≥5 papers published on the topic and were connected in 4 clusters (Figure 3). In both figures, the size of the circles represents the number of documents published by the author or country, and the thickness of the lines depicts the size of the collaboration between the authors or countries (Figures 2 and  3, respectively).

Co-occurrence
There were 290 high-frequency keywords, occurring ≥10 times, linked in 7 distinct clusters ( Figure 5). The cluster containing the most connected keywords was related to epidemiology and mortality. The second cluster included anxiety and depression, which were the two most occurring keywords, occurring 260 and 198 times, respectively ( Table 4). The most recent themes included vaccines, vaccination, machine learning, and online learning ( Figure 6).

Principal Findings
The COVID-19 pandemic has led to unprecedented global research activity. To the best of our knowledge, our bibliometric analysis is the first to quantify the published literature on COVID-19 research that has been conducted by countries in the EMR up until early 2022. We provided a descriptive evidence-based analysis of the published research, identifying leading countries and organizational affiliations, with visualizations of collaborations and evolution of COVID-19 research output. Research on COVID-19 in the region has been led by authors and institutions from Iran and KSA. There were multiple regional collaborative efforts; however, international collaboration is limited. Output focuses on COVID-19 epidemiology, mortality, and anxiety and depression. Recently, interest has been shifting more toward topics related to vaccination, machine learning, and online learning.
Our results show that KSA and Iran are leading COVID-19 research in the EMR in terms of the number of publications. These results are not surprising, as KSA also leads the COVID-19 publications in the Arab world, with a 35.65% share of research production [33], and Iran has been the largest contributor to biomedical and health research in the EMR, with a 39% share during 2004-2013 [34].
The high number of COVID-19 research publications in both countries, especially by their leading universities (Tehran University of Medical Sciences in Iran and King Saud University in KSA), is due to many factors. First, both countries have relatively large populations, 35 million and 84 million in 2020, respectively [35], and both were among the countries most affected by the pandemic in the region, especially Iran [8]. Since the beginning of the pandemic and until April 1, 2022, more than 7 million confirmed COVID-19 cases were reported in Iran [8], which is the highest number in the region. The rapid spread of the disease in Iran in several waves was associated with many unique political, social, cultural, economic, and religious dimensions [36], providing a stimulating environment for research activities. For example, religious tourism in Qom, a pilgrimage site in Iran, which continued during the pandemic, was a major source of spread to other Iranian cities, Pakistan, and KSA in the early stages of the pandemic [37][38][39]. In KSA, mass gathering events, such as Umrah and Hajj, which were put on hold during the pandemic, may have also contributed to the high number of publications, not because of the high risk of transmission, but because of the successful measures that KSA took to prevent the progression of the COVID-19 pandemic locally and globally [39][40][41].
Second, both countries have invested extensively in the needed resources for medical and scientific research along the years. Iran has built a large human resource capacity with its unique Iranian Ministry of Health and Medical Education and its mandatory and credit-based continuing education programs for physicians for relicensing, along with the increasing number of Iranian journals indexed in international databases [42][43][44]. KSA has also rapidly become a major player in scientific research in the EMR, as research and development (R&D) is 1 of the main pillars of the Saudi Vision 2030, with the specific goal of having at least 5 Saudi universities among the top 200 in the world by with fragile health systems, war torn, or with limited resources, such as Afghanistan, Djibouti, Somalia, Libya, and Syria, were less research-productive, although the number of cases in these countries may be higher than that officially reported and mandates further investigations.
Nevertheless, neither Iran nor KSA was reported to be at the top of the global COVID-19 literature in a recent global bibliometric analysis [47]. This adds to previous findings showing that the EMR share in global health research publications is low [34]. It is worth noting that Iran and KSA, like many other countries, faced various economic challenges during the pandemic that may have affected their research capacities. KSA was severely affected economically due to the radical decline in the global demand for oil, in parallel with the global economic recession, which was reflected in the sharp decline in oil prices during the first and second waves of the pandemic [48]. COVID-19 has also severely affected the Iranian economy, exacerbated by the concomitant international sanctions [49]. Generally, evidence shows that more support is needed for biomedical and health research in many of the EMR countries and institutions, especially those identified in our analysis as least productive.
International collaboration in COVID-19 research appears to be low in the EMR, as only 5.92% of the publications were associated with universities outside the EMR. The pandemic has shown that gaps in global cooperation, whether in research and information sharing, vaccine development and deployment, or travel policy, have affected the speed and equity of the global recovery [50]. In our study, the United States and the United Kingdom were at the center of collaboration and had the most extensive collaboration with EMR countries in COVID-19 research, similar to their role in COVID-19 research in the Arab world [33]. This could be due to the fact that both countries are among the top countries in scientific research [51] and specifically in global COVID-19 research [47]. The cross-border international flow of knowledge in the form cross-affiliation among researchers in the EMR with different international entities can be explained in several ways. First, the United States and the United Kingdom were by far the largest health donors to global health in 2019, with US $8.1 billion and US $2.9 billion, respectively [52], and when it comes to a donor-funded research project, most research funders require publication as a condition of a grant [53], triggering a cascade of cross-border research publications. Second, the United Kingdom and the United States have a network of research-intensive universities and scholarships, and both countries are hubs for international students, including those from the EMR, which could encourage cross-border collaboration. For example, the number of KSA students in the United Kingdom ranked eighth among international students from non-European Union countries, according to Higher Education Statistics Agency statistics for 2020/2021 [54]. In the United States, the number of KSA students ranked fourth among international students for the 2019-2020 academic year [55].
Most of the identified publications did not report a funding source, which could be due to limited funding opportunities or simply the likely dominance of cross-sectional studies and reviews, which are relatively less expensive, take less time to conduct, and require less infrastructure than, for example, clinical trials or prospective cohort studies [56]. The limited funding opportunities in the EMR may be related to the chronic underinvestment in research and development systems in the region, both financially and politically [57], exacerbated by the economic damage caused by the COVID-19 pandemic and the overreliance of some regional countries on external funding [58], although some major donors, such as the United Kingdom, have recently reduced their external funding [59]. Studies have also shown that research in the health care sector is not a top funding priority for governments, the private sector, or organizations in the EMR [57] and that R&D spending as a percentage of the gross domestic product (GDP) in the region is among the lowest in the world, despite significant differences in the GDP among member countries [58,60,61]. In EMR countries, R&D spending as a percentage of the GDP ranged from as low as 0.02% in Syria and Sudan in 2015 and 2019, respectively, to 1.3% in the UAE in 2018 compared to the global average of 2.27% in 2018 [58,61].
In our analysis of institutions funding COVID-19 research in the EMR, we found that 3 of the top 4 funders are from KSA, which is not surprising, given that KSA's spending on R&D as a percentage of the GDP was 0.82 in 2013 [61] and is likely to increase, given that KSA's GDP is still high [60], and the 2030 vision for R&D to increase universities' competitiveness and ranking will continue to create a more favorable environment for research [45].
Network analysis of coauthorship showed that the COVID-19 flow of information between EMR authors comes from a large and highly interconnected network with few isolated and small clusters. This indicates a high level of regional collaboration, either because of prepandemic collaboration and networking or because of governments and organizations' role in fostering collaboration since the onset of the pandemic. Regarding linkages, Iran had the most connections, consistent with another study showing that Iranian researchers have well-established international collaborations with leading countries in COVID-19 research [49].
Cultural differences could possibly have shaped COVID-19 research collaboration in EMR countries, which can be explained by Hofstede's theory of cultural dimensions [62]. A recent study analyzing research collaboration in the area of business and economics by 11 countries, including Iran from the EMR, before and after the onset of the pandemic, showed that Hofstede's uncertainty avoidance domain significantly influences the characteristics of research collaboration networks [63]. Uncertainty avoidance explains how cultures adjust to change and cope with uncertainty [62], and the pandemic has caused major changes and uncertainties at all levels of society worldwide [64]. However, using cross-national data from different EMR countries and interpreting them in the context of Hofstede's model of cultural dimensions will provide more accurate results and explore more details of the impact of culture on COVID-19 research collaboration in the EMR.
Evaluation of the co-occurrence network for keywords showed that the EMR COVID-19 research covers a wide range of COVID-19-related areas. The most prevalent keywords were anxiety and depression, which may be related to the global increase in both conditions during the pandemic [65,66], especially in the EMR, given the limited mental health services and widespread stigma [67]. In addition, the EMR is fertile ground for high exposure to misleading and unknown information, which may exacerbate the status of mental disorders [68]. Recently developing topics in EMR COVID-19 research include vaccines, immunization, machine learning, and online learning. However, future research needs to address other underresearched topics, such as ethics, infodemiology, and the human-animal interface, in line with EMR-specific challenges and the recent recommendation of the WHO Research and Innovation Report for COVID-19 [69][70][71].

Limitations
Our study was limited by factors that are inherent to its bibliometric nature, including the lack of ability to assess the quality of published documents; the fact that the number of citations may be misleading; and authors, countries, or institutions publishing early in 2020 are more likely to be cited than authors, countries, or institutions publishing in 2022. Additionally, within the EMR, a number of documents are published in Arabic, or other local languages, and in local journals and thus might not be indexed by Scopus and so would have been missed by this analysis. As is the case with other bibliometric studies, the results are dependent on the indexing of the databases, and even though Scopus was used for this bibliometric study, it should not be considered a comprehensive source of all published literature.

Conclusion
The majority of available research on COVID-19 in the EMR is published by authors and institutions from Iran and KSA, with a paucity of publications from other EMR countries. Studies from these other countries are crucial to identify context-specific public health challenges and determine culturally and socially appropriate interventions and best buys. There were multiple regional collaborative efforts; however, international collaboration was limited, even though a pandemic is an opportune time for international communities to join forces. Recently, and in line with global interest, regional interest has been shifting more toward topics related to vaccination, machine learning, and online learning. This gradual shift to relatively new concepts can be considered a major step in the forward move of the state of knowledge of the EMR. Understanding the current state of research is instrumental to future research production, and our study will inform regional research initiatives on emerging concepts, as well as opportunities for collaboration and funding.

Conflicts of Interest
None declared.