Internationalisation of general practice journals: a bibliometric analysis of the Science Citation Index database

Keywords: bibliometrics, general practice, primary health care, publishing, research.

Research is an important aspect of the development of medical specialities, and general practice medicine is no exception (Zarbailov et al. 2017). General practice and primary health care play a significant role in the functioning of health care (Hummers-Pradier et al. 2009). A solid research basis is essential for a strong and vibrant primary care system (Huas et al. 2019). In recent decades, general practice researchers around the world have been very productive in terms of publications (Mendis et al. 2010; Schneider et al. 2012; Vezyridis and Timmons 2016). However, wide disparities exist among different countries (Krztoń-Królewiecka et al. 2013). Several studies conducted with respect to primary care research output showed that from 1974 to 2017, the US and the UK had greatest the publication volume, and the UK, Canada and Australia had the greatest percentage of publications in primary care (Mendis et al. 2010; Schneider et al. 2012; Abdulmajeed et al. 2014; Hajjar et al. 2019).

Nonetheless, the number of publications alone is not an objective indicator of the quality of research (Sebo 2020). Scholars have advocated using the journal impact factor, the number of citations and research productivity to assess quality, but no consensus has been reached to date (Mårtensson et al. 2019; Sebo 2020). Even so, researchers in many countries have faced pressure because their performance is assessed according to the number of their publications in journals indexed in the Science Citation Index (SCI) database (Tijdink et al. 2014).

The trend for international authors to submit manuscripts to journals is not only a way to increase the global visibility of journals, but also a way for the research field to flourish (Loughborough et al. 2016). The continued growth and international positioning of journals is also inextricably linked to internationalisation (Elkins 2020). In bibliometric research, there are several metrics by which to evaluate the internationalisation of academic journals, such as the composition of the editorial board (Chen et al. 2003; Saxena et al. 2003; Dyachenko 2014), the geographical distribution of authors (Hart et al. 2007), readers/subscribers and those citing the publications (Tompkins et al. 2001) and the scope of the articles (Lin et al. 2014). Hart et al. (2007) investigated the proportion of international papers in gastroenterology and hepatology journals, with authors outside the countries where the journal were published classified as ‘international’. Researchers in other specialties have studied the internationalisation of their journals, but no such study has been conducted for general practice (Chen et al. 2003; Saadi and Mateen 2015; Schrock et al. 2016).

The features of general practice journals, especially their degree of internationalisation, should be of interest to general practice researchers worldwide, as well as to those who are concerned about the development of this speciality. Improving internationalisation can facilitate regional and international collaboration, influence policy-making processes and lead to the development of a global strategy to promote primary care under prevailing local conditions (van Weel and Kassai 2017; Clark et al. 2018). The aim of the present study was to analyse the volume of publication and internationalisation of general practice journals indexed in the SCI database.

In the past, there was no category for general practice in the SCI database. Then, in 2010, the distinct subject category of ‘primary health care’ was created, although it only included a meagre 14 journals at that time (van Weel 2011). The category was subsequently expanded to include 19 journals in 2019, with most of the journals being general practice journals.

We searched for ‘primary health care’ as the subject category in the 2019 InCites Journal Citation Reports – Science Edition and identified 19 journals. The Web of Science database was accessed through the Taipei Veterans General Hospital Library website on 30 July 2020, and the bibliographic records of these 19 journals were retrieved. Because the SCI database syntactically indexes the country of each author’s affiliation, we used the national distribution of authors’ affiliations to denote ‘internationalisation’. Because our aim was to gain an understanding of the most current internationalisation status of these journals, the time span of the records was limited from 1 January 2019 to 31 December 2019.

To understand the internationalisation of original articles, publications with a structured abstract (e.g. ‘articles’ and ‘reviews’) were included in the analysis. Notes, letters, editorials, news and meeting abstracts were excluded. The affiliations of authors from studies included were manually identified, whereas the countries of the publishers of the 19 journals were identified from their official websites.

An author was deemed to be ‘foreign’ if her/his affiliation was not located in the country in which the journal’s publisher was located. For UK journals, authors from England, Scotland, Wales or North Ireland were treated as domestic. For journals belonging to a multinational society, the definition of domestic was accordingly extended to a broader region. For example, for a journal published by a European Association, even if the journal is registered in the UK, all Europeans would be regarded as domestic authors. Similarly, for a journal owned by the Nordic Federation, authors from five Nordic countries were treated as domestic.

A paper was then categorised into one of three authorship types: purely domestic without foreign author(s), purely foreign without domestic author(s) and mixed type. The internationalisation ratio of a journal was defined as the quotient of the number of purely foreign and mixed-type articles divided by the total number of articles in a year.

Data were analysed using SPSS version 20.0. Chi-squared tests were used for group comparisons. Two-sided P < 0.05 was considered significant.

In all, 1573 papers published in 19 journals were indexed under the ‘primary health care’ subject category in the SCI database in 2019. Of these 1573 papers, 639 (40.6%) were authored or coauthored by foreign authors, with 493 (31.3%) papers authored by purely foreign authors and 146 (9.3%) authored by both domestic and foreign authors (Table 1). The degrees of internationalisation for the 19 journals varied widely. Family Practice had the highest degree of internationalisation, with 94.2% of its papers authored by foreign authors, followed by BMC Family Practice (88.2%) and Primary Health Care Research and Development (80.2%). In contrast, three of the journals had very low degrees of internationalisation: Primary Care (2.0%), American Family Physician (5.7%) and Journal of Family Practice (5.7%). Among these journals with low rates of internationalisation in 2019, Primary Care (US) published 49 articles with only one coauthored by authors from Ireland; American Family Physician (US) published 88 articles, with only one article authored by purely foreign authors (from Brazil) and four articles coauthored by authors from England, Canada and Japan; and the Journal of Family Practice (US) published 53 articles, with only one article authored by purely foreign authors (China) and two articles coauthored by authors from Canada, Ireland and Japan.

Table 1.  Internationalisation of authorships in general practice journals, 2019
Listed journals are ranked by impact factor

Of the 1573 SCI-indexed general practice papers, 86.4% (n = 1359) were published in four English-speaking countries: 32.8% in seven US journals, 34.8% in five UK journals, 12.5% in two Australian journals and 6.4% in one Canadian journal. Of the remaining articles, 13.6% were published in journals from European countries, excluding the UK (Table 2). One of the 19 journals, Atencion Primaria, focuses on work related to primary health care and is published in Spanish (https://www.journals.elsevier.com/atencion-primaria, accessed 20 November 2020). The UK journals had the highest degree of internationalisation (75.9%), whereas the degree of internationalisation of the North American journals was only 19.4%. There was a significant relationship between the country of publication and the degree of journal internationalisation (P < 0.05).

Table 2.  Internationalisation of general practice journals according to publisher’s country or region, 2019

The development of academic general practice has flourished in recent years (Lin et al. 2014). However, in the present study, based on the geographical distributions of authorship, the mean internationalisation of SCI-indexed general practice journals in 2019 was 40.6%. Table 1 shows a comparison of individual journals, which revealed considerable disparities in their degrees of internationalisation, with the highest rate being 94.2% (Family Practice) and the lowest being 2.0% (Primary Care).

Internationalisation in journals for other specialities has been evaluated by researchers worldwide. Hart et al. (2007) examined the degree of internationalisation of gastroenterology and hepatology journals in 2005 and found that the internationalisation ratio of all research articles was 67.6%. Chen et al. (2003) found that the percentage of international authors published in American Journal of Roentgenology in 2000–02 was 37%. Park et al. (2019) reported that the percentage of international authors in The Spine Journal increased from 17.8% in 2005 to 69.1% in 2015. Some studies have also revealed an increasing trend in internationalisation in journals in other specialities in past decades (Tompkins et al. 2001; Chen et al. 2003; Schrock et al. 2016). To date, the ideal ratio of internationalisation has not been comprehensively investigated. Hence, the lack of a gold standard for the degree of internationalisation is an issue that needs to be resolved.

Explanations for the low rate of internationalisation among general practice journals could include the editorial policies of the journals, reviewer bias, overall research production, the generalisation of research findings and researchers’ motivation to publish in a foreign language or in a particular journal (Langer et al. 2004; Goodyear-Smith et al. 2019). The scope and target audience of a journal are key determinants of the internationalisation ratio of general practice journals. For example, Family Practice (UK), which has a high percentage of internationalisation, orientates itself as an international journal of general practice, not one limited only to research conducted in developed countries (https://academic.oup.com/fampra/pages/About, accessed 20 November 2020). Moreover, Family Practice is not affiliated with any local or regional society. Another example is NPJ Primary Care Respiratory Medicine, which aims to inform and educate an international and multidisciplinary audience, including clinicians, respiratory specialists, respiratory physiotherapists, dieticians and nurses who take care of primary care patients with respiratory diseases (https://www.nature.com/npjpcrm/about/aims, accessed 8 July 2021). In contrast, internationalisation levels tend to be low among journals that belong to local societies. For example, American Family Physician (https://www.aafp.org/journals/afp/issues.html, accessed 20 November 2020) is per se an official publication of the American Academy of Family Physicians. Similarly, the Scandinavian Journal of Primary Health Care is published by the Nordic Federation of General Practitioners, which is owned by the five Nordic Colleges of General Practitioners (https://www.nfgp.org/, accessed 8 July 2021).

Researchers facing academic performance assessments may be more likely to submit their papers to high-ranked journals. Academic general practice developed later than other medical specialities, as reflected by the fact the ‘primary health care’ category was only recently included in the SCI database (van Weel 2011). As a result, the number of SCI-indexed general practice journals is still relatively limited, and the impact factors and ranks of these journals are not high compared as those covering other medical specialities (www.webofknowledge.com, accessed 20 November 2020). In addition, many GPs may submit their papers to journals covering subject categories other than primary health care (Sebo 2020). More importantly, the speciality of general practice emphasises psychosocial aspects of medical care that are deeply rooted in individual cultures (Farre and Rapley 2017); that is, the research themes and results of a study conducted in one country may not always apply to other countries.

Previous studies have examined possible solutions to the low rate of internationalisation among general practice journals (Klemenc-Ketiš and Kersnik 2014; Lisak 2014; Sugano 2015). First, it may be helpful to invite more people from abroad to join the editorial board, particularly those from countries near where the journal publisher is located that share similar customs and cultures (Sugano 2015). Second, English language editing resources available from publishers and other companies that can address language issues, making the manuscript more comprehensible and improving submission acceptance rates, especially for those authors for whom English is not their primary language (Lisak 2014). Third, international exchanges and seminars provide opportunities to build a global network, share ideas and investigate common problems in different regions (Klemenc-Ketiš and Kersnik 2014).

There was a statistically significant (P < 0.05) relationship between the publisher’s country and the degree of internationalisation of a journal. Most (86.4%) of the SCI-indexed general practice papers in 2019 were published in four English-speaking countries, namely the US, UK, Australia and Canada (Table 2). This implies English-speaking countries have an inherent advantage in publishing in international journals because they can easily dominate English writing. Significantly, the UK journals had the highest degree of internationalisation (75.9%), which may be attributed to international medical graduates accounting for approximately one-quarter of the General Medical Council register in the UK (Hashim 2017). In addition, 40% of UK medical students undertake medical electives, which are accomplished in the final year of study in a place and speciality of their interest, in developing countries. This may broaden cultural awareness and arouse greater concern for global health (Wallace and Webb 2014; Tiller and Jones 2018).

With the rise of international journals, debate has long prevailed as to the role of local journals. Local journals publish locally relevant research that is expected to disseminate knowledge and translate into practice and policy (Ofori-Adjei et al. 2006; Kristensen et al. 2015). In addition to local focus and appeal, local journals serve medical education roles by publishing secondary evidence or distilling guidelines and reviews into domestic language (Ofori-Adjei et al. 2006). However, local journals face challenges of fewer good-quality papers, fewer citations, less institutional support and a lower contribution to reputation building (Bodaghi et al. 2015; Seo et al. 2016). In addition, locals develop knowledge for domestic audiences, whereas internationalists generate knowledge for both local and international audiences, because increased international collaboration leads to higher publication rates (Kwiek 2020). With a vision of developing primary care research capacity, international collaboration can create practice-based evidence to support evidence-based primary care (Ponka et al. 2020).

This study has some limitations. First, we could only identify papers accepted and published by SCI-indexed general practice journals. We could not determine the full number of submissions or the countries of those submissions. This meant that we could not ascertain whether the low degree of internationalisation of a given journal was attributable to a low number of submissions from international authors or a low rate of acceptance and publication of submissions from international authors. Second, the internationalisation of general practice journals in this study may have been overestimated because we did not consider the close relationships among English-speaking countries or between the UK and other European countries. Third, this study was based on the SCI database, in which English is the dominant language in journal publications. Thus, our results cannot be extended to most general practice journals published around the world.

Although academic research in general practice has been developing vigorously in recent years, there is much room for improvement in the internationalisation of general practice journals in the SCI database. There are wide disparities in internationalisation among different countries and journals. Even though possible solutions to the low rate of internationalisation among general practice journals have been studied, their efficacy is uncertain. Further studies on the internationalisation of general practice journals may be needed.

The authors declare no conflicts of interest.

This study was supported by a grant (V109E-002-1) from Taipei Veterans General Hospital.