Scopus based bibliometric and scientometric analysis of occupational therapy publications from 2001 to 2020 [version 3; peer review: 1 approved]

Background: Occupational therapy (OT) is one of the allied health professions, with its first journals in 1920. The main objective of this study was to find out the publication trend in the field of OT research for the period of 2001-2020 using the principles of bibliometrics and scientometrics. Methods: The data was retrieved from Scopus from the past 20 years (2001-2020). VOSviewer software was used to find year-wise publications in OT-specific and non-OT-specific Journals along with top journals, countries, organisations, authors, cited articles, and highly used keywords. Results: There was a steady growth of OT articles over the past 20 years. Scopus indexes 16 OT-specific journals are identified. American Journal of Occupational Therapy, British Journal of Occupational Therapy Journal, Australian Occupational Therapy Journal, Scandinavian Journal of Occupational Therapy, and Canadian Journal of Occupational Therapy are the leading publications and citations. A comparison of OT-specific and non-OT journals inferred that the OT-specific papers are three times more published in non-OT journals. There is a trend in publishing multidisciplinary medical journals than OT journals. The US publishes the most articles

Any reports and responses or comments on the article can be found at the end of the article.

Introduction
Occupational therapy (OT) is one of the allied health professions practiced globally. Research and research publications are an integral part of the OT profession. "Archives of Occupational Therapy" was the first published OT journal published in 1922. In the 21 st century, most journals became online web-based after the invention of the internet . Thirty-nine journals have been published that contain "occupational therapy" in journaltitle available online. Thirty-two journals are published in English . These journals are core OT journal that mainly publishes OT-related articles. With the advancement in electronic databases and the online availability of periodicals, the newer bibliometrics and scientometrics methods of journal evaluation or evaluation of research measures are evolved (Meho and Yang, 2006). Citation analysis and content analysis are two commonly used methods of bibliometric analysis (Wallin 2005; T. . This method is proven to examine the impact or influence of published articles, journals, researchers, institutions, and countries. Further, this method evaluates microlevel performance, such as institution/university performance, to macro-level performance of a particular profession in research and country-wise research evaluation (Wallin 2005; T. . Content and citation analysis is a common practice in the bibliometrics of occupational therapy. In the past, most of the content analysis was performed for specific OT journals such as the American Journal of Occupational Therapy (AJOT) (Ottenbacher and Short, 1982;Ottenbacher and Petersen, 1985), Canadian Journal of Occupational Therapy (CJOT) (Ernest, 1983), Australian Occupational Therapy Journal (AOTJ) (Trevan-Hawke, 1986;Madill et al., 1989), British Journal of Occupational Therapy Journal (BJOT) (Cusick, 1995;Mountain 1997) and Occupational Therapy Journal of Research (OTJR) (G. T. Brown and Brown, 2005). Pearl et al. Most of the reported studies were based on the cited journal analysis, that is, analysis of what OT journal articles cited from other journals (Johnson and Leising, 1986;Roberts, 1992;Reed, 1999;Potter, 2010). Recently, Nowrouzi-Kia et al.
(2018) evaluated top publications with more than 100 citations in occupational therapy. This study was the first to uncover the highest annual citation rates, randomized control trials, literature reviews, and cross-sectional studies in occupational therapy. Ziviani and colleagues (1984) analyzed the content and citation of three OT journals, AJOT, BOTJ, and AOTJ. It was the first bibliometric study examining both content and authority in the OT field. Recently, bibliometric studies have become a popular method to know the publication trends, find researchers/authors on a particular topic, and identify journals where prospective authors select to publish their manuscripts. This study type helps identify a specific profession's publication landscape (MacDermid et al., 2015). These data are also used as criteria for research promotion, researcher incentivizing, grant allocations, and policymaking. Further, these data can be used to benchmark faculty, department, institute, or research organization (T. Brown et al., 2019).
Several studies were reported on the performance of OT researchers. Those studies evaluate OT authors' publication output regarding their content and citation impact. All analyses were conducted for OT authors from specific geographical locations such as Canada (MacDermid et al., 2015), Brazil (Folha et al., 2017), the United Kingdom (T. , and Australia (T. . Recently one study reported the bibliometric analysis for authors of western countries and Asian countries. It includes a total of nine countries among them, the United States (US), Canada, the United Kingdom (UK), and Australia were included as countries of the west, and Hong Kong, Japan, Taiwan, Singapore, and South Korea were included as Asian countries (Man et al., 2019). Among those few publications, it was observed that OT authors had published many publications in non-OT journals (MacDermid et al., 2015;. Many highly cited OT papers are also published in non-OT-specific journals (Man et al., 2019). Another study on OT publication in non-OT journals from 2004 to 2015 found that publication in non-OT journals increased by 173% (Folha et al., 2019). Because there has been an explosion of new online electronic journals with open-access options for publications, publishing is a multidisciplinary type of journal that has become a global trend among prospective authors in any discipline.

REVISED Amendments from Version 2
There was error in writing abbreviation of low-and middle-income countries (LMIC). The term LIMC is replaced with LMIC in the whole of revised manuscript.
Any further responses from the reviewers can be found at the end of the article Recently all bibliometric studies in the field of OT used Journal Citation Report (JCR) matrices and Journal Impact Factor (JIF) for their analysis provided by Web of Science (WOS) (T. MacDermid et al., 2015;Gutman et al., 2017;Folha et al., 2017;Man et al., 2019). A simple mathematical formula calculates JIF. The total citations will be in the numerator and the citable item as the denominator for the previous two years. The denominator creates confusion because the authors cite anything from the available literature. Journals do not label the publications as citable items in their author guidance. Neither prevents any item from citing (Fernandez-Llimos, 2018). Similar to JIF, Scopus CiteScore also uses citation analysis for ranking journals. The Scopus database that provides CiteScore is among the two databases accepted worldwide. The other is the Web of Science (WOS), which provides JIF (T. Brown and Gutman, 2019;Fernandez-Llimos 2018;Sau, 2020).
The CiteScore metrics released in 2017 offer access to citing and citation articles. It computes all published materials as citable articles (Fernandez-Llimos, 2018). However, there was concern about citable items such as "notes", "letter to the editor", "editorials", "erratum", and "retracted", and there will be many unidentified items. This error has been fixed in the recently updated version of CiteScore metrics 2020, where they have fixed for the "articles", "reviews", "conference papers", "book chapters", and "data papers" published. Scopus CiteScore has more excellent comprehensive coverage of published research data (Fernandez-Llimos, 2018;Sau, 2020). Both JIF and CiteScore calculations are based on citations received by a journal in a given period for the published citable item in that duration (Fernandez-Llimos, 2018;Sau, 2020). Significant and strong correlations were found between JIF and CiteScore in recent studies comparing 14-OT journals published in English (T. Brown and Gutman, 2019). Hence, both JIF and CiteScore for measuring journal quality are used in our current study.
Thus, it indicates a requirement for a comprehensive bibliometric evaluation of OT literature using accurate and accessible matrices, which can be used to understand the global prospect of OT literature. These reasons prompted us to conduct a detailed bibliometric and scientometric analysis of research output in OT using the Scopus database. Hence, the following research questions were proposed.

Ethical approval
This study was conducted based on data retrieved from Scopus between 2001 and 2020. Due to no human subjects' involvement, ethical approval was not required.

Source of data
Scopus database was used for the study. Scopus is one of the international scientific committee's best bibliographic databases (Fernandez-Llimos, 2018;Sau, 2020). Scopus has covered 70 million items and 1.4 billion cited references since 1970, making it the most extensive research publication database. Scopus bibliometric information is expensively used in the scientometric analysis. Citations vary between databases and it is based on the number of journals indexed in that database. We selected Scopus for this study because it includes more OT-specific journals compared to WOS and overall indexed more journals compared to WOS. The citation count was calculated within the database. Due to extensive coverage (Fernandez-Llimos, 2018;Sau, 2020), Scopus has a more citable item and a probability of getting more citations to count.

Data retrieval strategies
The search was conducted in December 2021 and was limited to scientific articles and reviews on OT published from 2001 to 2020. We used the TITLE-ABS-KEY function using the advanced search option of the Scopus database. Keywords such as "Occupational Therapy" and "Occupational Therapist" (T. Nowrouzi-Kia et al., 2018; was used along with multiple Boolean operators such as "AND", and "OR" to retrieve the maximum number of relevant articles. We excluded non-peer-reviewed documents such as "short surveys", "conference papers", "editorials", "notes", "letters", "books", "book chapters", "erratum", and "retracted papers" (T. Nowrouzi-Kia et al., 2018).

Scientometric analysis
The VOSviewer ® , a free software, was used. Microsoft Excel 2016® was used for all the figures and scientometric calculations.
Scientometric data was retrieved from Scopus using a comma-separated file (CSV) with complete information. The full details, such as citation, bibliographical, abstract, keywords, funding, and other information, were downloaded as a CSV file to analyse the data using VOSviewer.
Identification of year-wise publication trends All journal titles were screened to identify the OT-specific journals from the Scopus database. The journals with titles that contained the "Occupational therapy" word were categorised as OT-specific journals (Folha et al., 2019). Additionally, we verified all journals' title, which contains either of these three words "Ergotherapia", "Occupational Therapy", or "OT" and confirmed the organisation that published those journals. Further, the occupational therapy organisation or association publishing journals were included in the OT-specific journal list. The remaining journals were categorised as non-OT-specific journals. Then year-wise CSV file was imported into VOSviewer to generate year-wise publication of OT-specific, non-OT-specific information. We used VOSviewer software citation per source function for this purpose. VOSviewer uses "source" instead of "journal" in the software. Citation per publication is used to generate ten-year journal publication output. Then the information is imported into an Excel file to compute year-wise OT publications, year-wise publications in OT-specific journals, and year-wise publications in non-OT-specific journals. The cumulative publications for all three categories were calculated separately to identify the growth trend of OT publications.

Identifications of top journals, authors, countries, and organisations
The VOSviewer was used to categorise the top twenty journal sources, countries, and 20 organisations based on publication numbers. Similarly, we used twenty-year files to identify the top twenty authors who published at least 25 articles and the top 20 individual articles based on the number of citations. We referred to analysis using CiteScore 2020 from the Scopus database and Journal Citation Report 2020 from WoS Master Journal List to tabulate the Scopus CiteScore and JIF for the top-twenty journals.

Country collaboration network and authors' keyword visualisation
A network map on international collaboration about OT publication and author keywords visualisation was generated using VOSviewer software.

Results
The overall OT publication trends revealed a steady growth of (7.72%) each year. In OT-specific and non-OT-specific journals publication, growth was 0.29% and 7.44%, respectively ( Figure 1). Approximately one-fourth (24.30%) of articles were published in OT-specific journals in the last twenty years, and three-fourths (75.69%) were published in non-OT-specific journals. From 2001 to 2020, OT publications increased by 165.14 % (n = 1544). Similarly, the overall OT publication in OT-specific journals increased by 14.84 % (n=57). In contrast, non-OT-specific journals overall publication increased by 269.85% (n =1487). The current analysis found that 14 of the top 20 are OT-specific journals, which published one-fourth of OT articles from 2001 to 2020 (Table 1). Among those journals, the AJOT received more citations (n = 27770). However, the Archives of Physical Medicine and Rehabilitation (APMR) received the highest citation per document (37.27). The CiteScore of those 20 journals ranges from a minimum of zero to a maximum of 5.7. Fifteen of the top 20 journals have journal impact factors (JIF) or citation indicators (JCI). Six journals were indexed in WOS's Emerging Sources Citation Index (ESCI). In contrast, two journals were not listed in WOS. Impact factors for those 15 journals were rage from 0.03 to 3.966.
Among the top OT publishing countries, the US published the most articles (n= 9517), followed by the UK, Australia, Canada, and Germany ( Table 2). The Netherlands received the highest citation per document (41.84), followed by Denmark, France, Italy, and Norway. Furthermore, data revealed that the US, UK, Australia, and Australia have solid international collaboration among the best 20 countries published in the past 20 years (Figure 3).  Table 4. Further, the top 20 highly cited      publications are listed in Table 5. The article published in 2017 received 2309 citations and was the most cited article in the Lancet. We observed a negative correlation (-0.14278) when comparing citations with the total year after publication.
Our study enlisted 20 author keywords commonly used in OT literature. Overall, the best 20 keywords were identified, with at least 50 occurrences in the OT publications ( Figure 4).   Discussion Over the last twenty years, overall OT publication output steadily increased, indicating the profession's growth in research, and its dissemination is not significantly influencing the modern era of digitalization. Publication in non-OT-specific journals is three times more compared to OT-specific journals. A similar trend was also observed in recent studies (T. Brown et al., 2019;Gutman et al., 2017;Folha et al., 2017;Folha et al., 2019). This increase may reflect the growth of research activities in OT across the globe (Folha et al., 2019). Occupational therapy works along with medical and other allied health professionals. It allows occupational therapists to work as a part of a multidisciplinary team. Similarly, it opens the door to interdisciplinary research opportunities. That multidisciplinary research is used to publish in either medical-related or multidisciplinary journals. Those journals have higher JIF than OT journals (T. Brown et al., 2019;Gutman et al., 2017;Folha et al., 2017;Folha et al., 2019). A recent study found that non-OTspecific journals have three times more JIF than OT-specific journals (Folha et al., 2019). These multidisciplinary and medical-oriented journals have more readers than OT-specific journals, giving more visibility and increasing the chance of receiving more citations. It may also encourage OT researchers to publish more in non-OT-specific journals (T. Brown journals are well-known in OT and have a long publishing history (T. Brown and Gutman, 2019). AJOT, BJOT, AOTJ, and CJOT were published by prominent national organizations of OT and recognised by the global OT community (T. Brown and Gutman, 2019). These journals' publication frequencies are more, and they publish more articles per issue than other journals. Among these 16 OT journals, two journals, BrJOT and IrJOT, are new in the Scopus database and have data from the past three years. Hence, those journals will have a lower number of articles. According to the previous study, Scopus-indexed OT-specific journal numbers were 14 (T. Brown and Gutman, 2019). Among these two new journals, BrJOT published many articles in three years and placed among the best 20 journals that published OT-specific articles. OT-specific journals published one-third of total OT publications. Several OT-specific journals are published globally and have an online publication (Cruz et al., 2019) but are not included in the Scopus database. Including all those OT-specific Journals might change the citation matrices of occupational therapy journals because occupational therapy journals tend to get more citations than OT-specific journals.
Out of the 20 top journals that produced the maximum number of articles in occupational therapy, the first eight are OT-specific journals. There is a normal phenomenon because these journals published OT-specific articles. ErgoR received more minor citations among those ten OT-specific journals due to its publication language. This is the only journal listed in the top twenty published in Germany. All other periodicals are published in the English language. AJOT, BJOT, AOTJ, SJOT, and CJOT received more citations in terms of overall citation. It may be due to their large volume of publications (T. Brown and Gutman, 2019;. AJOT, BJOT, AOTJ, SJOT, and CJOT received more than ten citations per document. Archives of Physical Medicine and Rehabilitation received more citations per document than any other journal in the top 20. It has the highest CiteScore and JIF. Among the OT-specific journals, POTP had the highest CiteScore (2.9), and AJOT had the highest JIF (2.246).
The US, Australia, the UK, Canada, Germany, Sweden, Netherlands, and Italy are the top most country in terms of the number of published articles, and this is different from that of the previous study (T. Brown and Gutman 2019;. The US published approximately one-third of total OT publications (n=9517). Other study findings also revealed that the US is the top OT article (T. . The US received many citations (216810) for the published documents. The US is where OT originated from and has a good education and research system in OT, which may be the reason for the huge number of publications. The US is also the host country for many OT-specific and non-OT-specific journals. This could also be a reason for the huge number of publications in occupational therapy. However, citation per document is more minor for US articles than in the Netherlands, Denmark, Italy, France, and Norway, with very few publications compared to the US. It may be due to the lack of availability of OT-specific journals in their geographic location, which triggered them to publish their research in non-OT-specific journals with higher quality than OT-specific journals. The finding also suggested that high-income countries (HIC) have produced more publications than low-and middle-income countries (LMIC). It may be due to the lack of human resources, financial support, and infrastructure for research and policies in LMIC (Prvu et al., 2019;Regalado et al., 2023).
We found that the US, Australia, the UK, and Canada are the leading countries in international collaborations in OT publications. The size of the level and circle indicate the weightage of collaborative activity. A bigger size represents more collaborations, and a smaller size means fewer collaborations. The line between two nodes and their densities represents their link and strength ( Figure 3). Color expressed different cluster levels based on international collaborations (Van Eck and Waltman, 2019). The finding suggests that HIC courtiers have more international collaborations with more HIC than LMIC. It may be the similarity of disease burden among HICs compared to LMIC.
Among the top 20 organizations, 18 are universities or entities of a university. All the top 20 organizations are from HICs. These universities are mainly placed in the urban set up with all necessary support, such as research culture among academicians, well-equipped libraries with academic resources, affiliated hospitals, client participation, internal funding, and grant office supports (T. Brown et al., 2019). This is favorable for more research output, which may cause the university domination's top twenty organization list. Our study found only one organization in this top list: research institutes from Denmark and one occupational therapy association from the USA. Though the US alone produced many OT articles, only five organizations are listed in the top 20, compared to the US, Canada (n=6), and Australia (n=5). Despite having a high volume of publications, less representation of US organizations in the top list was also observed in previous studies (T. Man et al., 2019). Other countries that are listed in top organizations are Denmark (n=1), Iran (n=12), Sweden (n=1), and Taiwan (n=1).
Australia (n=11) has the highest representation in the top author list, followed by Canada (n=1), Sweden (n=4), the US (n =3), and Japan (n=1 Of the top-twenty highly cited articles from 2001 to 2020, four papers were published in 2008, two in 2005, 2006, 2009, and 2012, and one in 2002, 2003, 2004, 2007, 2013, 2014, 2016, and 2017. Our study does not find any relation with year after with more citations. Our findings opposed previous studies' findings, which observe a long citation window helps an article gather total citations . Most of the highly cited papers are published in non-OT-specific journals. Most highly cited studies were published in non-OT-specific journals like earlier studies. Those non-OT journals are generally high-quality medical journals . It may happen due to the increased pressure to publish in journals with good impact factors or CiteScore. The quality of journals provides more visibility, which helps the author get more readership and citations because it helps in a grant application and promotion . VOSviewer density visualisation (Figure 4), by default, uses blue, green, and yellow to represent the density visualisation (Van Eck and Waltman, 2019). The yellow indicates the number of items in a point's neighborhood and the neighboring items' higher weights. Out of the total best 20 keywords, "occupational therapy", "rehabilitation," "stroke", "physical therapy", and "activities of daily living" are five common keywords used more frequently in OT research, which are in yellow ( Figure 4). Occupational therapy and rehabilitation are commonly used as keywords because these two words are identical terms for the occupational therapy profession. The authors used these two words to make their research visible in the electronic search. Stoke is probably the oldest and strongest research field where occupational therapists work globally. Physical therapy is a common allied health profession occupational therapists may collaborate and publish together. Activities of daily living are one of the core practice areas of the occupational therapist.

Conclusion
Our study shows the dominance of HICs over LMICs in research production. This might be due to multiple factors unfavorable for LMICs to conduct research, which may be due to a lack of resources or knowledge translations published in a specific language (Prvu et al., 2019;Regalado et al., 2023). This scientometric research outcome gives a glimpse of OT research worldwide. It suggests further research in OT to discover why the difference in research output between HICs and LMICs and the different factors that influence occupational therapy research in HICs and LMICs because these information might be essential for developing country-specific research priorities or developing strategies in the field of occupational therapy. Similarly, other disease-specific scientometric research must be conducted to determine top researchers and countries' collaborative institutions. This study retrieved scientometric information from the Scopus database and listed 25 journals with their CiteScore and Journal Impact Factor. It was observed that OT articles were published three times more in non-OT-specific journals, indicating that OT research significantly overlaps with other disciplines of medicine. Hence, the investigation should not conclude a literature search with only OT-specific journals. The SJOT is one of the core OT journals but is listed under the subject category of "public health, environmental and occupational health" rather than the category of "occupational therapy".
Key points 1) From the Scopus database, twenty Journals were identified, which published a maximum number of occupational therapy articles.
2) Scopus database indexed a total of 16 OT-specific journals. Scopus database also included two OT practice magazines in their database.
3) VOSviewer software is an open-access tool that can be used as a cost-effective method for any scientometric analysis.
4) BJOT, AJOT, AOTJ, SJOT, and CJOT are the leading occupational therapy journals in the number of published items.

Data availability
Underlying data Mendeley Data: Scopus Based Occupational Therapy Research (2001Research ( -2020. https://doi.org/10.17632/yp7xjg4zs3.2 (Sau and Nayak, 2022) This project contains the following underlying data: Data File "20 best source which published maximum number of article" contains analysis of the data obtained from Scopus for 2001-2020. The analysis includes 20 best sources, 20 best cited documents, 20 best authors, 20 best countries, 20 best organisations, 20 highest used keywords, year-wise publication analysis, and OT-specific journal details.
Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

The benefits of publishing with F1000Research:
Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com