Bibliometric analysis of road traffic injuries research in the Gulf Cooperation Council region

Background: Despite governmental interventions, the Gulf Cooperation Council (GCC) region continues to experience higher road traffic crash and fatality rates relative to Western nations. This trend suggests a potential disconnect between Road Traffic Injuries (RTI) research and the mitigation measures put in place. Method: Here, we present an in-depth bibliometric analysis to obtain a comprehensive understanding of RTI research in the GCC region. The Web of Science database was used to search and retrieve the relevant articles during the period of 1981-2019. Results: The volume of RTI research increased from 2015–2019, suggesting an increased focus on traffic safety in the GCC region. Saudi Arabia had the highest RTI research productivity level (126 publications); Bahrain had the lowest (7 publications). Inconsistent with its low publication volume, Hammad Medical Corps of Qatar had the highest citation impact score of 16.33. Global collaboration for RTI research was highest between Saudi Arabia and the United States. The most prevalent publication journal for the region was Accident Analysis and Prevention. The most common keywords were “ road traffic accidents” and “ road traffic injuries”; terms such as “ mobile phones”, “ pedestrian safety”, “ pedestrians”, and “ distracted driving” were least common. In the five most productive GCC nations with respect to RTI research (Saudi Arabia, United Arab Emirates, Qatar, Kuwait, and Oman), researchers tended to publish works related to road traffic safety in traffic safety-oriented journals. Conclusions: The quantity and quality of RTI publications in GCC is insufficient to meet the increasing related public health and economic burden in the region. The trends among publication volumes, citations, and impact were inconsistent. There is a lack of research collaboration among the institutions. Most of the research related to RTI is being conducted by researchers with a medical background. Research focusing on pedestrians, cyclists and road user behavior is also inadequate.


Introduction
Road traffic injuries (RTI) account for 30% of all deaths worldwide and are the major cause of death among people 15-29 years old 1 . Figure 1 depicts the leading causes of death expected in 2030 by the World Health Organization (WHO) 2 . RTI was ranked as the eighth contributor to the total global deaths in 2004; by 2030, RTI is expected to rank as the fifth highest contributor 2, 3 The Gulf Cooperation Council (GCC) region (Saudi Arabia, Oman, the United Arab Emirates, Kuwait, Qatar, and Bahrain) experiences substantially higher road traffic crash and fatality rates relative to Western nations. Across 15 developed European nations, the recorded road traffic fatality rate has decreased from 13.5 deaths per 100,000 populations in the 1980s to 5.5 deaths per 100,000 populations today. In contrast, the road traffic fatality rate in the GCC region has remained relatively constant at 23 deaths per 100,000 population over this same period 4 . Table 1 compares the 2013 road traffic fatality rates per 100,000 population and per 100,000 registered vehicles for each of the GCC nations, as well as the United Kingdom and the United States 5 . In the GCC region, Saudi Arabia and Oman had the highest fatality rates per 100,000 population, followed by Kuwait, Qatar, the United Arab Emirates (UAE), and Bahrain 5 . The population-based fatality rate in the UAE (10.9 deaths per 100,000 population) was comparable to that in the United States (10.6 deaths per 100,000 population). However, the fatality rate based on vehicle ownership in the UAE (38.2 deaths per 100,000 registered vehicles) was significantly higher than that in the United States (12.9 deaths per 100,000 registered vehicles). The United Kingdom had the lowest fatality rates in terms of both population (2.9 deaths per 100,000 population) and vehicle ownership (5.1 deaths per 100,000 registered vehicles) 5 .
As demonstrated in other nations, road traffic crashes and fatalities can be mitigated through a focused program of RTI research that identifies and supports implementation of strategies designed to reduce the magnitude and severity of road traffic crashes and fatalities. Historical RTI research has been conducted in Western nations with a focus on relating aggressive driving behavior and offences to traffic crashes [6][7][8] . Early studies showed that more than 75% of all crashes were caused by driver behavior [6][7][8][9][10][11] . Although driver behavior clearly constitutes an important risk factor, researchers today caution against trying to identify a single crash cause. Socioeconomic and demographic factors, such as age, sex, marital status, education, training, experience, way of life, emotional status, fatigue, reaction time, vision, vigilance, and driving speed may also affect crash occurrence and should be considered as risk factors in RTI research 12 .
The high and sustained number of road traffic crashes and fatalities in the GCC region relative to other comparable Western nations poses serious public health and economic

Amendments from Version 1
Changes have been made in the light of reviewer's comments for the improvement of the original manuscript. Figure 2 Flow Diagram for the literature review process was added in Methodology Section. Any further responses from the reviewers can be found at the end of the article REVISED challenges and has been well documented in previous studies [13][14][15] . These challenges are exacerbated by the GCC region's rapidly expanding roadway network and increasing vehicle ownership levels. Unlike demonstrated efforts in Western nations to reduce the magnitude and severity of road traffic crashes and fatalities based on a focused program of research, the high and sustained number of road traffic crashes and fatalities in the GCC region suggests a potential disconnect between local RTI research and any mitigation measures put in place.
To address this potential disconnect, we performed a bibliometric analysis of RTI research in the GCC region using publication data from the Web of Science indexing and abstracting database for 1981-2019. We considered research productivity, institutional and individual authorship, bibliographic coupling and global collaboration, publication journals and publications, and keywords and associated topical trends for the GCC region. In addition, we performed a three-factor analysis using GCC nation, publication journal, and keyword parameters to obtain a more comprehensive understanding of RTI research in the GCC region. The results from this study can inform governmental authorities in the GCC region of research investment needs and academic researchers of research deficiencies and impactful publishing venues. An informed and collective focus on RTI research in the GCC region will help to ensure broader public safety.

Methods
In this study, we performed a bibliometric analysis of RTI research in the GCC region. Bibliometric analysis is a statistical method that collects, analyzes, and extracts various metrics related to research productivity, institutional and individual authorship, publication journals, publications, keywords, and more from published literature. We used data from the Web of Science (WOS) indexing and abstracting database. The WOS database was chosen to extract the relevant data on RTI. Web of Science is the most authentic and reliable indexing and abstracting global database of scholarly literature. Additionally, it also provides full bibliometric data with a simple extraction process, which is suitable for comprehensive bibliometric analysis. The following search query was used in the topic field of the advance search option of the Web of Science Core collection database on November 13, 2019.To get maximum relevant records we used all keywords related to RTI. We searched using the following search terms in the topic field (that is title, keywords and abstract) of the WOS database. These search terms were selected based on literature review and author keyword analysis in different databases. The authors tried to include all search terms for retrieval of entire spectrum of the literature related to study objectives (see Figure 2). This initial query returned 571 publications. The authors did not apply any filter related to languages and time period to achieve a comprehensive view of RTI research. We refined this initial query by excluding editorial materials (4), notes (2), meeting abstracts (5), letters (2), or corrections (1) as they are not peer reviewed. Research articles, conference papers, review papers, book chapters, and books were included in this study. The complete bibliographic information (title, abstract, author, date of publication, source of publication, keywords, citation count etc.), of the remaining publication records were exported to MS Excel (Microsoft 365) for scrutinizing of these records. Two authors read the title and abstracts of all the records. The third author again repeated this process to ensure the relevancy and reliability of the data. This practice helped the authors to discard 245 irrelevant records. Finally, a total of 312 records were selected for data analysis. The final dataset for this study included 312 publication records comprising 230 journal articles and 82 conference papers 16 . Data analysis was performed using a combination of spreadsheet and bibliometric analysis software packages such as VOSViewer (version 1.6.15) and Biblioshiny (R Package). The author keyword analysis and bibliographic coupling was done with the help of VOSViewer. Global collaboration, topical trends and three factor analysis were generated by using Biblioshiny.

Limitations of Study
In this study, we performed a bibliometric analysis of RTI research in the GCC region using data from the Web of Science indexing and abstracting database for 1981-2019. This data was not independently verified to determine if each of the publication records considered for analysis originated from the GCC region. Other databases such as Scopus, PubMed, or Google Scholar may produce a different set of publication records using similar search criteria. Although this comparison was beyond the scope of this study, future work may attempt to verify this study's findings using data from these alternate sources.  Publications originating in Saudi Arabia were cited 890 times and had a citation impact of 7.06. Despite lower publication volumes, publications originating in the UAE and Qatar had higher citation impact (7.79 and 7.76, respectively). Table 3 lists the most productive institutions for RTI research in the GCC region. Four institutions were in Saudi Arabia, three were in the UAE, two were in Qatar, and one was in Kuwait.

Bibliographic coupling among nations
Bibliographic coupling is a measure of subject matter commonality among different publications and occurs when two publications reference a common third publication. Figure  Publications that are closely related in content can be classified into clusters; the connections between the clusters can be described using quantitative network indicators. A visualization  The UAE University had the highest RTI research productivity with 40 publications cited 545 times with a citation impact of 13.64. King Saud University and Qatar University also had a high number of publications (37 and 29 publications, respectively). The Hamad Medical Corporation and Al Ain Hospital had just 15 and 12 publications, respectively, but their citation impact ranked highest (16.33 and 14.58, respectively). Table 4 lists the most productive individual authors for RTI research in the GCC region. Only three authors had 10 or more of the clustering network can then be generated based on research publication characteristics. Publications spanning different time periods can be clustered and any sub clusters can be merged into a unified visual view 17,18 . In this study, nations with three or more publications were included; 23 of 45 nations met this criterion. Bibliographic couplings in the RTI research publications originating in each of these 23 countries were classified into five clusters and visually displayed using the VOSViewer software package.

Individual authorship
Global collaboration among nations Figure 5 depicts the historic global collaboration in RTI research among nations. In total, 232 instances (i.e., publications) of global collaboration involving GCC nations were identified. These instances ranged from a single collaboration between two nations to more than 11 repeated collaborations between two nations. The highest levels of collaboration have occurred among Saudi Arabia and the United States (11 instances), Qatar and the United States (9 instances), Egypt and   the UAE (8 instances), Saudi Arabia and Tunisia (8 instances), and Qatar and the United Kingdom (7 instances).

Authorship patterns
As a supplemental investigation, we considered the number of individual authors on each RTI research publication. Figure 6 shows the number of publications as a function of the number of authors for publications originating in the GCC region. Select publications had more than 10 individual authors, although this was relatively infrequent. Most RTI research publications originating in the GCC region had no more than five authors; the highest number of publications (74) had three authors.  Keywords that were listed four or more times were included; 33 keywords met this criterion. The most common keywords in RTI research publications were traffic safety (24 occurrences), road traffic injuries (17 occurrences), and road safety (17 occurrences). Terms such as pedestrian safety (5 occurrences), pedestrian (5 occurrences), mobile phone (5 occurrences), driver behavior (4 occurrences), in-jury prevention (4 occurrences), and distracted driving (4 occurrences) were least common.

Topical trends
Supplemental to the keyword analysis results presented previously, we considered topical trends in RTI research publications originating in the GCC region from 1992 to 2018. Figure 8 depicts these results. In 1992-2007, RTI research publications were limited to just six topics: Saudi Arabia, hospital, Riyadh, region, accident, and mortality. In 2008, the breadth of topics began to increase. In 2009-2019, the most common RTI research publication topics included road, traffic, safety, and accidents.
Three-factor analysis using nation, journal, and keyword parameters As a final step in this study, we performed a three-factor analysis using GCC nation, publication journal, and keyword parameters to obtain a more comprehensive understanding of  RTI research in the GCC region. Figure 9 shows the diagram of published literature on RTI Gulf focusing on the relationship among three factors including top keywords, journals and country. Figure 9 presents the results from this analysis.
RTI research originating in the five most productive GCC nations (Saudi Arabia, the UAE, Qatar, Kuwait, and Oman) was most commonly published in just four journals: the International Journal of Injury Control and Safety Promotion, Accident Analysis and Prevention, Traffic Injury Prevention, and the Saudi Medical Journal. The focus of this research was tied to just three keywords: traffic safety, driving, and road safety.

Discussion
Research related to RTI and road traffic awareness first appeared outside of the GCC region in 1926 and 1966, respectively 23 . Comparatively, the first publication related to RTI in the GCC region did not appear until 1981, confirming a lag in RTI research activity among GCC nations relative to other developed nations.
Following that first publication in 1981, RTI research productivity gradually increased in the GCC region but remained relatively low until 2015-2019, when the volume of RTI research substantially increased. A similar trend was reported in India; publication volumes related to road traffic injuries in India increased significantly after 2010 24 . These collective results suggest an increased recent focus on traffic safety in the GCC region by policymakers, planners, and academic researchers. This increased research productivity can also be attributed to an increased number of print and online publication journals, increased attention to this topic within the scientific community, increased networking and collaboration capabilities, and an increased focus from governmental and specialized agencies such as World Health Organization (WHO).
Disparities in RTI research productivity were observed between the GCC region and other developed nations as well as within the GCC region. The GCC nations with the highest number of publications included Saudi Arabia (126 publications), the UAE (84 publications), and Qatar (50 publications). This finding is consistent with results previously reported by Abdo et al. 23 who found that Saudi Arabia had the highest number of publications related to road traffic safety and road traffic awareness relative to all other GCC nations. Saudi Arabia has a higher population and more research institutions than other GCC nations.
Trends among publication volumes, citations, and impact were inconsistent. The 126 publications originating in Saudi Arabia were cited 890 times and had a citation impact of 7.06. Despite lower publication volumes (84 and 50 publications, respectively), publications originating in the UAE and Qatar had higher citation impact (7.79 and 7.76, respectively).
Regarding the publishing institutions, the UAE University had the highest RTI research productivity with 40 publications, which were cited 545 times with a citation impact of 13.64. King Saud University and Qatar University also had a high number of publications (37 and 29 publications, respectively).
Again, trends among publication volumes, citations, and impact were inconsistent. Although the Hamad Medical Corporation and the Al Ain Hospital had just 15 and 12 publications, respectively, the citation impact for these publications ranked  Research, by its nature, is a collaborative process. In any bibliometric analysis, the level of research collaboration is an important index to assess the current status of research in a specific field. Considering research collaboration by nation reveals both the degree of international communication as well as the most influential nations in a particular research field. In this study, the highest level of collaboration occurred among Saudi Arabia and the United States (11 instances). This finding is consistent with a previous study's findings that found that the highest numbers of collaborative publications in the field of Physical Sciences (1980-2014) were produced through partnerships in Saudi Arabia and the United States (23.31%) and Saudi Arabia and Egypt (22.95%) 25 . Similarly, the highest numbers of collaborative publications related to Health Sciences in Saudi Arabia were produced through partnerships between Egypt (16.5%) and the United States (16.3%) 26 .
These collective findings highlight the strong collaborative and intellectual ties Saudi Arabia maintains with respective academic communities in the United States and Egypt. However, these ties should be further strengthened not only among the GCC nations but also with other developed countries such as Australia and Germany. Australia, the United States and Germany are currently leading the world in terms of best practices, regulations and road testing of connected and automated vehicles (CAVs) 27 , which represent the future of transportation and traffic engineering and safety research.
We also considered the number of individual authors on each RTI research publication. Select publications had more than 10 individual authors, although this was relatively infrequent. Most RTI research publications originating in the GCC region had no more than five authors; 60 publications had a single author, 54 publications had two authors, and 74 publications had three authors. The trend depicted in Figure 6 shows a decline in publication collaborations involving four or more authors in the GCC region. Instead, publication collaborations involving a smaller number of authors has been favored. Also, the number of single author publications was high. Efforts to improve collaboration among individual authors from different departments and institutions would improve both the quality and impact of research originating in the GCC region.
Among the most prevalent publication journals for RTI research in the GCC region, Accident Analysis and Prevention had the highest number of publications (25), number of citations (533), and citation impact (21.32). Of the 20 most commonly cited RTI research publications originating in the GCC region, eight appeared in Accident Analysis and Prevention.
The most commonly cited RTI research publications originating in the GCC region related to traffic safety, and to a lesser extent, traffic awareness. A number of the commonly cited RTI research publications specifically addressed medical issues associated with these topics. This finding is consistent with findings previously reported by 24. Of the 20 most commonly cited RTI research publications originating in the GCC region (Table 6), only two publications considered pedestrians (the most vulnerable road users): one published in Accident Analysis and Prevention 21 and one published in Traffic Injury Prevention 28 . None of these publications considered more contemporary or emerging topics, such as the role of mobile phones in road traffic injuries.
Consistent with the publication-related findings, the most commonly listed keywords in RTI research publications originating in the GCC region were road traffic accidents (41 occurrences) and road traffic injuries (40 occurrences). Terms such as traffic safety (27 occurrences), road safety (20 occurrences), and road traffic crashes (19 occurrences) were also commonly listed. Terms such as mobile phones (5 occurrences), pedestrian safety (5 occurrences), pedestrians (5 occurrences), and distracted driving (4 occurrences) were least common.
To detect changes over time in RTI research focus in the GCC region, we considered topical trends from 1992 to 2018. In 1992-2007, RTI research publications were focused on the medical aspects of the traffic safety; hospital and mortality were common topics in RTI research publications. In 2008, the breadth of topics began to increase. In 2009-2019, the most common RTI research publication topics included road, traffic, safety, and accidents. This trend can be attributed to the increasing RTI burden in the GCC region, which experiences ever-increasing road traffic crashes and deaths on the roads each year despite the development of both in-vehicle and highway-based safety system technologies. Traffic safety awareness among governmental traffic planning department staff and academic researchers has increased due to improved international data collection and sharing practices and the periodic Global Status Reports on Road Safety in 2013 1 , 2015 5 , and 2018 29 published by the WHO.
This aspect of the investigation not only revealed the historic RTI research focus in the GCC region, but also revealed important deficiencies that can guide future research. For example, public safety in the GCC region may benefit from future research focused on pedestrian behavior and safety or the effects of mobile phone usage and distracted driving.

Future research directions
More broadly, future RTI research should focus on developing preventive actions to increase and promote traffic safety rather than only addressing the medical aspects of traffic injuries in the GCC region. This bibliometric analysis has revealed minimal research related to human behavior factors and pedestrians and cyclists who are most vulnerable for road traffic injuries in the GCC region. Traffic safety researchers and practitioners in the region should focus on these research areas, which show significant publication potential. Researchers should also focus on improving communications between departments and institutions to facilitate sharing and usage of updated data among researchers. This latter effort will also encourage collaboration among authors. As noted previously, efforts to improve collaboration among individual authors from different departments and institutions would improve both the quality and impact of research originating in the GCC region.

Conclusions
To address the potential, disconnect between RTI research and any traffic safety mitigation measures put in place in the GCC region we performed a bibliometric analysis using publication data from the Web of Science database. We considered research productivity, institutional and individual authorship, bibliographic coupling and global collaboration, publication journals and publications, and keywords and associated topical trends for the GCC region. In addition, we performed a three-factor analysis using GCC nation, publication journal, and keyword parameters to obtain a more comprehensive understanding of RTI research in the GCC region. Non-motorized transport users such as pedestrians and cyclists are the most neglected group in the road traffic research in GCC nation despite of having higher fatality rates. The researchers should focus on the safety impacts of the pedestrians and other non-motorized road users who are the most volatile and susceptible to road accidents. The researches related to road user behavior and on the usage of cell phones and seatbelts are also not on the satisfactory level in GCC. It is of utmost importance that the data sharing and collaboration between the institutes should be improved so that the quality and impact of the future RTI researches can be improved because of experienced teamwork and professional feedbacks. This will also help to improve the current trend of researchers and institutions who are currently working in silos rather than in collaboration with the traffic practitioners, planners and professionals.
Across several of the bibliometric parameters, trends among publication volumes, citations, and impact were inconsistent. For example, a high volume of publications did not always result in a high number of citations or a high citation impact.
This study demonstrated the merit of applying bibliometric analysis to guide decision-making related to RTI research.
The collective results of this study indicated that-despite the recent increase in RTI research productivity in the GCC regionthe quantity and quality of RTI research publications originating in the GCC nations is insufficient to meet the increasing burden of RTI in the region. More effort is needed to increase and promote traffic safety. The collective results from this study can inform governmental authorities and policy makers in the GCC region of research investment needs and support cost-effective resource allocation. In addition, the results from this study can inform academic researchers of research deficiencies and impactful publishing venues. An informed and collective focus on RTI research in the GCC region will help to ensure broader public safety.

Open Peer Review
those demonstrations and how you have concluded here that there is a disconnection in GCC.
Results: Sub-section : Three factor analysis... Why 3 factors are considered?
○ I see the importance of nations and keywords but not able to comprehend inclusion of journal (what exactly it will reflect it is not clear to me). ○ Discussion: Para 1: Could you please discuss here what was the reason why before 1981 there wasn't any research study related to RTI? Is it because the RTI problem was not that significant or was no attention given by academics or no funding was available or there wasn't any research culture or academic performance was not graded based on research activity...which is clearly the case now.
Para 4: why articles from KSA have low impact factor compared to UAE and Qatar? Could you elaborate on the possible reasons for this? Para 6: medical journals usually have way higher impact factors and have large audience compared to non-medical journals (planning and engineering-based journals). I do not agree on the conclusion you made here that universities are not producing high-quality research just based on the citations impact.
Para 8: My comment here is similar to what I mentioned above, authors need to categories, you cannot ignore the content and make your conclusion on citations.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound? Partly

If applicable, is the statistical analysis and its interpretation appropriate? Partly
Are all the source data underlying the results available to ensure full reproducibility? Yes

Are the conclusions drawn adequately supported by the results? Partly
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Transport Planning, Road Safety, Transport Modelling, Big data.

Uneb Gazder
Department of Civil Engineering, University of Bahrain, Zallaq, Bahrain This paper presents a comprehensive bibliographic analysis of the research related to road safety in GCC countries. All important trends have been analyzed in detail and their possible reasons have been discussed. The authors have also highlighted important issues related to past trends and suggested future improvements. The following corrections are required to be made in the manuscript before accepting it for indexing: Figure 1 is not cited before its appearance.
○ Data in table 1 should be updated.

○
There are repetitions in the discussion section, for e.g., an explanation of bibliographic coupling and collaboration that have already been done in the previous sections. They should be removed from the discussion.