Introduction
Bibliometric indicators are useful for assessing research performance; however, this is valid when the indicators are right, sophisticated, current, included expert knowledge, and evaluated and used with care (1). The importance of research which is indicated by the citation count is quantified by using citation data in bibliometric studies (2). The Institute for Scientific Information (ISI) has been compiling the most relevant bibliometric information from published scientific articles since 1945, then the Science Citation index (SCI), a specific tool for measuring citations, was available since 1962 (3). Today, in bibliometric studies, the citation indexes created by Thomson Reuters, particularly the Web of Science (WoS) and its predecessor, the SCI, are commonly used databases (1).
Many bibliometric analysis was published in medicine including the topics of oral and maxillofacial surgery (4,5), psychiatry (6), plastic and reconstructive surgery (7), neurosurgery (8), neurophysiology (9), neurosciences (10,11), orthopedic surgery (12), urology (13), pulmonary diseases (14), public health (15). Bibliometric analysis in dentistry has been performed in such areas endodontics (2,16-20), implant (21-23), paediatric dentistry (24), periodontology (25,26), oral medicine and radiology or radiology (27-29), orthodontics (30) and in such topics dental traumatology (31,32), orofacial pain (33), dental education (34). Some citation studies were based on articles published in several dentistry journals (3,35-37), articles published in a single journal (17,38) or articles published by authors in a single country (39).
Analysis of research results may help get information on the development of policies related to scientific and technological aspects in dentistry, especially in developing countries (e.g. China, Thailand, South Korea, Brazil) and in the EU which is currently undergoing a major reorganization (40). According to ISI database from 1999 to 2003, an increase in dentistry-related studies was monitored in developing countries, such as Brazil, Turkey, China, Thailand, South Korea, Jordan and Iran. Moreover, the average number of citations for each document was reported as 1.27 and it was reported that Turkey was in the 10 countries with most published articles from 1999 to 2003 (40). The data from SCImago showed that Turkey was the 4th country in Europe and 8th country in the world in terms of the number of dental articles published between 1996 and 2018. Also, a total 550 citable article from Turkey was published in 2018 with 0.36 citation per article (totally 218 citations) (41).
In the literature, there is no article specifically on the top-cited papers in dentistry published by authors affiliated to the institutions in Turkey. Another study analyzed the characteristics of the top 100 articles which are the most cited on dentistry which are published in scientific journals with international collaboration whose coauthor(s) affiliated to the institutions in Brazil (39). In this study, it was stated that such analysis could help provide the current status of Brazilian dentistry regarding the most cited dental articles, as well as the areas and topics which attract international attention. So, country-based citation analysis studies will provide a panorama in terms of the scientific studies of that country in that field and will provide scientist with guiding data in terms of future planned studies. Also, this study will provide a wider perspective to evaluate the studies that are in the field of dentistry. Therefore, in the present study, we aimed to evaluate and analysis the 100 top-cited articles published by first author affiliated to Turkey institutions.
Materials and Methods
A search about dental articles with authors from Turkey was performed on 23 March 2020 on the WoS (Clarivate Analytics Co., Philadelphia, PA), which is an online platform (http://www.webofknowledge.com) providing bibliographic data on scientific materials.
In order to investigate the publications of the authors addressed in Turkish institutions in both in dental journals and other journals containing articles related to dentistry, as the first step, “All Database” was selected from “Select a Database” option. Following the election of “Advanced Search” option, without any year or language restrictions, the proposed query set [CU =(TURKEY)] was scanned and this search resulted in 658,322 articles (n=658,322). Finally, we used “Dentistry, Oral Surgery & Medicine” as the “Research Area” category filter. With these limitations, 18,243 articles are obtained as a result (n=18,243).
In the present study, the articles without the first authors affiliated to a Turkish institution and publications of the source types such as “books”, “book series”, “conference proceedings”, “trade publications” and “undefined” were excluded. The selection of the articles was performed by two independent researchers (M.D. and F.K.). The articles were listed from the highest to the lowest based on the number of citations received and the 100 most cited articles were identified (Table 1). Accordingly, as for articles which have same number of citations, the more current one was ranked higher.
The 100 articles receiving the most citations were analyzed with regard to citation count, title of the article, the year of publication, the number of authors, journal name, institutions, collaborations type, type of study, distribution of articles to subfield of Dentistry and subfield topic, study design of clinical articles, thematic field and level of evidence of clinical study. Any disagreement between two researchers on article analysis was resolved through discussion and consensus.
The journal names are listed based on the number of the articles which are top-cited, but journals with the same number of articles are ranked based on the Journal Impact Factor (JIF) 2018 which is provided from the Journal of Citation Reports (https://jcr.clarivate.com) (Table 2).
The main institution was indicated by the address of the first author in Turkey. The institutions and the addresses of co-authors were also considered to detect the type of institutional collaboration without considering the departments. Thus, institutional collaboration type was considered as individual institution when all contributing authors were affiliated to the same research institution regardless of the departments. The other institutional collaboration types were classified as “multi-institutional collaboration” if multiple institutions within the Turkey joined the study and as “international collaborations” if there were institutes from different countries (18,20). If the first author is working in multiple institutions addressing Turkey or there is more than one country cooperation in one article, each one of these institutions and countries were counted.
The study types included basic science, clinical research, and review. A study classified as a basic study when it included experiments on extracted human teeth, animal, dental plaque, microorganism, saliva, dental materials or cells (18). While determining the study field, the full text of all the articles in the list was read and discussed until the consensus was achieved by the two researchers to determine the study types. In addition, an abstract of each study in PubMed has been reached and the Medical Subject Headings (Medical Subject Headings terms) given for this study field have been considered. The main subject of the articles was categorized as prosthetic dentistry, endodontics, periodontology, orthodontics, restorative dentistry, oral and maxillofacial surgery, pedodontics, oral and maxillofacial radiology and oral implantology according to the department of which the first author is affiliated. If first author was not affiliated to dentistry, the department of co-author which was affiliated to dentistry was based on. Where no author was affiliated to dentistry, the main subject of the articles was determined according to topic of article. After this, each article for each main subject was classified according to its own subtopics which were investigated in article (Table 3).
The previously described the study design of clinical articles were used as follows: randomized controlled trial (RCT) controlled trial, systematic review of RCT, non-randomized experimental study, systematic review of cross-sectional study among consecutive presenting patients, cross-sectional study among consecutive presenting patients, cross-sectional study among non-consecutive patients, case-control study, case report, and cross sectional study. Afterwards, the level of evidence (level I to V) was defined for the clinical articles according to the type of research question (intervention 1, diagnosis, prognosis and etiology) (Table 4) (42).
Statistical Analysis
Statistical analysis for frequency of descriptive measures was performed using SPSS version 21 (IBMCorp, Armonk, NY).
Results
The 101 top-cited articles are listed according to the citation count in a descending order in Table 1. The last four articles have the same number of citations (91 citations); therefore, 101 articles were listed. The highest top cited article had 351 citations and it is in in vitro study on the fracture resistance of endodontically-treated teeth which were restored with different post systems. This study related to Prosthetic Dentistry field and was published 2002. Least cited four articles had 91 citations and one of them is in vivo study which is related to orthognathic surgery, three of them are in vitro studies on the cytotoxicity of root canal sealers, the strength of translucent fiber post bond to dentin and the biomechanical effects of rapid maxillary expansion, respectively. These studies related to Prosthetic Dentistry and Orthodontics field and they were published 2006 and 1998, respectively. The 101 most top-cited articles received 12,577 citations in total and the mean citation count for each article was 125 (Table 1).
Journals and The Year of Publication
The top-cited most-cited articles were published in 35 different journals whose publication language is English. 13 journals published three or more top articles. Two of the top-cited articles were published in four journals, and one of them was published in 18 journals. The JIF of 35 journals was between 43.07 and 0.785. Journal of Prosthetic Dentistry published the highest number of top-cited articles (n=17) and “Journal of Endodontics” was the second journal (n=13), followed by “Journal of Clinical Periodontology” (n=5) and “Journal of Oral Rehabilitation” (n=5) (Table 2).
The 101 top-cited articles were published between 2016 and 1981. Fourteen of them was published between 1981 and 2000, 78 between 2001 and 2010 and 9 between 2011 and 2016. In 2004, the highest number of top-cited articles (n=15) was published, which followed by the years of 2002 and 2005 (n=13), and 2007 and 2008 (n=8) (Figure 1). Journal of Prosthetic Dentistry was the first to publish a top-cited article back in 1981. It was related to Prosthetic Dentistry field which was about “causes of fracture of acrylic resin dentures”. The newest article was published in Nature in 2016. It was related Restorative Dentistry field which was about “ablation-cooled material removal with ultrafast bursts of pulses” (Table 1).
Authors and Institutions of Origin
A total of 342 unique authors contributed to top 101 most-cited-articles. The number of authors in top-cited articles was between 1 and 12. Four articles had single author. Twenty-two articles were attributed to two authors, 18 articles to 3 authors, 24 articles to 4 authors, and 15 articles to five authors. The number author of 16 articles was between 6 and 9 (Table 5).
The number of institutions indicated by the address of the first author in Turkey was 22 (Table 6). 13 of the 22 institutions had two or more top-cited articles which were published. Among the 13 institutions, Hacettepe University had the highest number with 20 top-cited articles (2,393 citations), followed by Ege University with 11 top-cited articles (1,557 citations), Marmara University with 10 articles (1,142 citations) and Selçuk University with 10 articles (1,098 citations). Considering all the authors that contributed to an article, 45 articles were produced by individual institutions, 27 by multi-institutional collaboration within Turkey and 29 by international collaborations. Considering the collaborations made internationally, those with the institutions in the United States had the highest number of published top-cited article (13 articles), followed by those in Finland with three articles, Sweden with three articles and United Kingdom with two articles (Table 7).
Type of Study, Field of Study, and Study Design, Level of Evidence and Type of Research Question of The Clinical Articles
Among 101 top-cited articles, basic science had the highest number of articles with 54 articles. 40 articles focused on clinical research and 8 articles were review. Whenever an article included 2 or 3 topics which were separately defined, it was scored accordingly in 2 or 3 categories (Table 4) (18).
Most of the topics of the 54 top-cited basic science articles were implant with 6 articles, followed by post and/or core materials with 5 articles and root canal sealer with 5 articles. Major topics of interest of 40 clinical research were implant with 4 articles, followed by gingival crevicular or chronic periodontitis with 4 articles. The majority of the topics of 8 top-cited review articles were implant with two articles (Table 3). Prosthetic Dentistry was the field with the highest number of top-cited articles (n=34) followed by Periodontology with 16 articles and Endodontics with 14 articles. Major topics of articles in different dentistry field were shown Table 3.
RCT was the type of study design with the highest number of articles (21 articles) in clinical experience, followed by cross-sectional study with 10 articles. The most frequent type of research question was classified as intervention with 22 articles, followed by diagnosis with 12 articles. Level of evidence was classified as having level II in 27 articles and level II in 7 articles.
Discussion
The number of citations to an article is not necessary for the quality of that article. However, citation analysis studies that include all disciplines of dentistry can provide clues for author and topics contributing to major development in dentistry (3). Citation analysis studies on different topics such as country-based are used as a tool to evaluate research performance from both institutional and individual point of the process, including funding and hiring decisions, determination of research needs for resource allocation, and investment in research facilities of field or specialty (34). Citation analyses indices showed an increasing trend. Moreover, findings of citation analyses studies would be beneficial for the editors and publishers of dental journals as a comparative criterion for measuring success and trend of their respective journals as well as providing a self-evaluation for the dental community (37). Therefore, this study evaluated and analyzed the 101 articles which are top-cited and were published by the authors affiliated to the institutions in Turkey.
In the present study, the numbers of citations were between 351 and 91, and the mean citation count for each article was 125. In contrast to our results, Feijoo et al. (3) reported the citation numbers were between 2050 and 326 for the top 100 most-cited articles which were published in dental journals. In another study, it was reported that the citation received by dental journals was 97,081 in 2003 and 233,232 in 2012. Also, the total number of dental journals which made citations of the articles significantly increased from 46 in 2003 to 83 in 2012 (37). On the other hand, 101 top-cited articles received a total 12,577 in our study. This difference may be due to our study being country-based and involving a single country. Studies involving different countries can enable more topics to be studied and collaborate and can provide a wider perspective on study topics. Another study country-based citation analysis which included the top 100 articles which were the most-cited articles and were published in international dental journals with at least one coauthor affiliated to an institutions in Brazil showed citation numbers between 124 and 657 (39). The difference between this and our study may be due to difference in study design. In our study, the main institution was indicated by the address of the first author in Turkey. However, in this study, it based on at least one coauthor affiliated to Brazil. Therefore, this may have led to greater collaboration opportunities in top cited studies. As a result, it may increase the citation number of top-cited articles compared to our results.
In our study, the majority of the top 101 most-cited articles (71 articles) were published in the journals with a JIF of >2, which included ≥2 top-cited articles. Besides, 20 of the 71 articles were published in the journals with a JIF of ≥3, which is relatively high, and included ≥2 top-cited articles in the field of dentistry. In contrast to our results, it was found that 59 of the top 100 most-cited articles were published in the journals with a JIF of >2, and 2 or more top-cited articles were published in 58 of the 59 journals published (3). This difference may be due to date of publication of this study which published in 2014. Because the JIF of journals may change depending on time. Difference may also be due to the difference between the journals where top 101 cited articles are published. In accordance with this assumption, it was found that the JIF of 10 dental journals significantly increased from 2003 to 2012, but a wide variation in the magnitude of this rise was observed (37). In accordance with our finding, in a study among the top 100 most-cited articles by the authors from Brazil, it was revealed that 65 of the 100 articles which are the most-cited were published in the journals with a JIF of >2, which included ≥2 top-cited articles (39). It was reported that it is known fact that authors consider a JIF as the most important determinant while selecting a journal for publishing their research, on the other hand, high JIF journals draw attention of high quality articles, thereby attracting high quality papers (27). However, in the present study, 20% of the most-cited articles were published in those journals with relatively higher JIFs (3 or more). In agreement with our findings, no significant association has been found between the JIF of a journal and the number of top-cited articles (20,27).
In our study, 92 of the 101 (~92%) top-cited articles were published before the year of 2010. In accordance with our findings, in other studies, the majority of the articles belonged to a decade or more before the publication of the studies (3,18,20,22,27,30,32,39). In contrast to our results, some studies have reported that those articles which are the most-cited have been published in the last 10 years (2,16,24). It was stated that the oldest articles usually had much more time ahead to be cited than those which have been recently published, without taking their scientific impact account, thereby carrying the risk of excluding the most recent influential articles (27). The total number of citations to an article can only increase over time (39). It was demonstrated that an article requires a minimum of six to 15 years to receive a sufficient number of citations following its publication and become a citations classic (22). In the present study, only 9 top-cited articles were published after the year of 2010 which means those articles have had the last 10 years to be cited. Similarly to our finding, it was reported that the possible reason is the limited amount of time for important studies to receive higher number of citations, or that there is now a higher tendency to submit qualified articles to journals which have the highest JIFs and that those journals are included in other categories of the Journal Citation Report (3,43).
Co-authorship is the resulting close collaboration between or among authors who intensively cooperate during the improving of research and take responsibility for the study content through joint signature, so that, in any situation, they can present and defend the original idea of the research (44). In our study, 24 articles had 4 authors. The number of top-articles with 2-6 authors is 89. Similar with our results, another study which is based on at least one co-author affiliated to an institution in Brazil found 79 top-cited articles had 2-6 authors (39). A study on scientific collaboration in research in Brazil revealed that the most recurrent number of is the 6-author authorship, which accounts for about 76% of its scientific production when added to 5 and 7 or more authors (44). In our study, approximately half of top-cited articles (45 articles) showed a single-center collaboration in the form of individual institutions. Also, international collaboration (29 articles) has secondly the highest citation. The most collaborated country was United States followed by Finland and Sweden. In accordance with our findings, it was found that the country with the most international collaboration was the United States, followed by Finland (39). The explanation for the situation might be the fact that the better the economy a country ranks enables it to produce the higher quantity and quality of biomedical publications (28,45). However, in our studies, the number of a single-center collaboration among top-cited articles was found higher than multi-institutional and international collaboration. It was suggested that young researchers may focus on broad range collaboration of relevant researchers from different disciplines to create a study with high impact/quality (28). Also, a study showed that international collaboration increased the rates of citation which an article received (46). However, in our study, international collaboration with 3600 citations (29% of total citation) did not effect on citation rates. Besides, 18 of 22 (81.8%) institutions with 91 articles are public and 4 of 22 articles (18.2%) are private institutions with 9 articles. This finding agreement with another study in which one co-author affiliated to an institution in Brazil found that of the 138 institutions coauthoring the articles, 73.9% were public and 26.1% were private institutions (39).
In our study, more than half of the top-cited articles (54 articles) were in basic science, followed by clinical research (40 articles) and review (8 articles). In contrast to our study, Feijoo at al. (3) found plurality of clinical research (66%) over basic research (34%). Another study, which is based on minimum one co-author affiliated to an institution in Brazil, found that 75 top-cited articles included epidemiological, clinical, and basic research, and 25 top-cited articles were reviews including those both narrative and systematic (39). In the present study, most of the top-cited articles were on the field of Prosthetic Dentistry with 34 articles and it was followed by the fields of Periodontology and Endodontics with 16 articles and 14 articles, respectively. Partially agreement our results, a study with co-authors from Brazil revealed that the most frequent subjects addressed in 100 top-cited articles were Endodontics, Periodontology, and Dental Materials (61% of the sample). A study including the top 100 most-cited articles in the field of dentistry found that the majority of research was on the field of Periodontology (43 articles), and it was followed by the fields of Implantology and Adhesive Restorations with significantly less number of material (11 articles and 8 articles, respectively) (3). In another study related to the citation rates on the articles in Scopus which were co-authored by the institutions in Brazil, it was showed that Endodontics had the highest total citations followed by Multidisciplinary and Dental Materials (46). In our study, implant for Prosthetic Dentistry, root canal sealer for Endodontics and, periodontal disease and molecular mechanism for Periodontology were main topic of top-cited articles. On the other hand, for Orthodontics, Restorative Dentistry, Oral and Maxillofacial Surgery, and Pedodontics, there was not predominant topic. In another other 100 top-cited articles studies on one specific area, microbial etiology of periodontal diseases for Periodontology (25), endodontic microbiology for Endodontics (18,20) and implants for Orthodontics (30) were major topics.
In our study, the most common study design of the articles in clinical research were RCTs (21 articles) followed by cross-sectional studies (10 articles). Level of evidence was II at most (27 articles), followed by III (7 articles). In contrast to our findings, the cross-sectional study (24), case series (3,18,21,22,30) or case reports (16) was reported as the most common clinical study design. This difference may be due to difference in study design as most of these studies performed on one area of Dentistry different from our study. Also, it was reported that majority of top-cited clinical experience articles had low level evidence such as level III evidence (20) and IV or V (18,22,27,30) than our study. It was stated that this might indicate relatively easier method of conducting simpler studies (30). In accordance with level of evidence, most of clinical researches were RCTs in our study. It was explained that studies place into a so-called hierarchy of evidence, with systematic reviews and meta-analyses of RCTs at the top contributing to the highest level of evidence, followed by RCTs, non-RCTs, cohort studies, case-control studies, crossover studies, cross-sectional studies, case studies, and expert opinions and uncontrolled studies or opinion at the bottom (36,42,47).
Conclusion
In our study, the top 101 most-cited articles had lower citation when compared to other 100 top-cited articles due to the first author affiliated to Turkey institutions. Journal of Prosthetic Dentistry published the highest number of top-cited articles and Prosthetic Dentistry was the field on which the highest number of articles conducted. A total of 92% of the top-cited articles were published before 2010. The number of top articles with 2-6 authors is 89. More than half of the top-cited articles were in the field of basic science. Most of the topics of the 101 top-cited articles were implant. The most frequently used study design was RCT for the articles in clinical experience. Level of evidence most of clinical researches were level II. Compared to the literature, it is promising that the number of clinical trials and level evidence is high. However, it is necessary to increase the number of international collaborations to boost the citation count of article.
Ethics
Ethics Committee Approval: This article does not contain any studies with human participants or animals performed by any of the authors.
Informed Consent: For this type of study, formal consent is not required.
Peer-review: Externally and internally peerreviewed.
Authorship Contributions
Concept: M.D., S.T., Design: M.D., F.K., Data Collection or Processing: F.K, M.B., Writing: M.D., F.K. Critical review: S.T., N.T., Supervision: M.D.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study received no financial support.
