Globally, the number of scholarly journal publications in the biomedical and health science field has risen sharply in recent years [1]. The number of journals indexed in Medline increased from 3,823 in 1995 to 5,623 in 2016. Also, biomedical journals published in Korea are greatly increasing [2]; between 2000 and 2017, the number of journals indexed by the Korean Association of Medical Journal Editors (KAMJE) increased from 57 to 256 [3]. However, the citation performance of journals published in Korea is relatively low [4] and has remained stagnant over decades [5].

A citation is often used as a criterion to evaluate an article's quality and even as a criterion to assess the research competence of an individual or a research team [4]. Therefore, researchers try to publish their study results in a journal that has a high citation rate. Also, editors of journals make various efforts to increase the citations of their journals. Some journals permit open access to enhance readers' access and attract more readers [6]. In non-English-speaking countries, such as South Korea, some journals are published in English to extend the scope of readers and to increase the probability of an article to be cited [7]. In addition, editors of journals try to enlist their journals in international databases [8] because the enlistment of a journal in international databases not only guarantees, to some extent, the quality of the journal but also makes an article easily attract readers' attention. Hence, enlistment in international databases may help a journal to acquire trust from researchers and thereby increase the probability of citation.

Over the past few decades, many bibliometric measures that reflect academic journals' scientific quality and influence were developed [9]. Traditionally, the major source of citation data has been the Web of Science (WoS), which reports impact factors (IFs) Eigenfactor scores, and article influence scores (AISs) for the world's leading journals enlisted in databases such as the Science Citation Index (SCI) [10]. Another major database is Scopus, providing SCImago journal rank (SJR), source normalized impact per paper (SNIP), h-index, and impact index [10]. Also, the Korean government has established the Korea Citation Index (KCI) to analyze the citation rates of domestic research [11]. The KCI reports Korea Citation Index impact factors (KCI IFs) and Web of Science-Korea Citation Index integrated impact factors (KCI-WoS IFs).

Although the Korean government aims to enhance the quality of domestic journals through various national policies, no empirical study has been conducted about the effectiveness of those policies. Also, little is known about the factors influencing the citation of journals in biomedical and health sciences. Investigation of the factors of citation rates may contribute to the spread of domestic research works and relevant future policies. Therefore, the objectives of the present study were to compare citation rates according to database enlistment and publication characteristics and to identify the factors influencing journal citation rates.

This study was conducted to analyze the factors influencing the citation rates of biomedical and health science journals published in Korea. The results show that enlistment in an international database enhances the citation rate. Most of the citation rates of the journals enlisted in prominent international databases, such as SCI and Medline, were higher than those of the journals that were not enlisted. This result could be explained by the fact that researchers in the biomedical field try to cite the articles of journals enlisted in prominent international databases to support the validity and legitimacy of their studies. In addition, journals enlisted in prominent international databases are more likely to be searched, attracting more international readers [6].

The multiple regression analyses confirmed the results that the key factor influencing international citation rates was enlistment in prominent international databases. Enlistment in SCI was found to be effective in increasing SNIP and SJR, even after other variables were adjusted. Enlistment in PMC was found to be effective in increasing SJR, even after other variables were adjusted. On the contrary, enlistment in KCI was found to be a factor decreasing IF. Only three journals were enlisted in SCI but not enlisted in KCI, and the IF of these journals was high because the major targets of these journals were international readers.

Enlistment in Medline was found to influence AIS and KCI-WoS-IF, even after other variables were adjusted. The effect of enlistment in Medline may be based on not only the fact that Medline is a prominent international database but also the fact that Medline is one of the databases most frequently used by researchers of medical science [14]. Moreover, the articles of the journals enlisted in Medline undergo MeSH (Medical Subject Headings) indexing by MeSH indexing experts, which may increase the probability to be searched by researchers. Researchers want to rapidly find desired articles by using the MeSH indexing languages as search keywords to increase search specificity [1516]. In studies such as systematic reviews, MeSH indexing is often used for search keywords to increase comprehensiveness [17]. However, the causal relation between MeSH indexing and citation rates should be clarified in further studies.

In addition, open-access journals showed higher IF, SNIP, and SJR scores than non-open-access journals. This result is consistent with those of previous studies that found that open-access journals are easily accessed by researchers and are more likely to be cited [18]. The Scopus citation rates, SNIP, and SJR scores of the journals published only in English were higher than those of other journals, which is consistent with the findings of previous studies [619]. This suggests that journals published in English are viewed by international readers and thus may be cited more frequently.

However, the KCI IFs were lower for journals published only in English. This could be explained by the fact that journals published in Korean are easier for Korean readers to read; thus, the domestic citation rates of these journals were higher. This explains the low KCI IFs of journals enlisted in SCI, Embase, and PMC as well as open-access journals, which is opposite to other citation rates. Most of the journals enlisted in these databases are published in English, and publication in English is a prerequisite for an open-access journal to be enlisted in one of these databases. Journals published in English showed lower KCI IFs than journals published in Korean. These results suggest that publishing a journal in Korean is advantageous to enhance domestic circulation in Korea. In addition, journals having higher publication frequency showed lower KCI IFs. Therefore, to improve the quality of a journal published in Korean, publishing high-quality articles may help to increase domestic citation rates, such as KCI IFs, rather than simply increasing the number of articles.

Various citation indices were considered in the analysis performed in the present study. These citation indices are calculated by using different equations and have different implications [20]. Therefore, the predictors of individual citation indices are different. For example, the SCI citation index, the popularity index, IF, and the prestige index, AIS, were found to have different predictors. Although IF is the most widely known citation index, and it is extensively applied due to its simple calculation, IF has many limitations, for example, lack of quality assessment of citations, English language bias, and influence of self-citation [21]. Also, other indices have their own advantages and disadvantages [22]. Therefore, in the evaluation of the quality of a journal, it is important to understand the calculation equations of individual citation rates and to utilize a citation index appropriate for the evaluation purpose, rather than relying on a single citation index.

The present study had following limitations. First, since this was a cross-sectional study, it may have been limited in identifying a causal relation between database enlistment and citation rates. A further longitudinal study needs to be conducted to verify the direct impact of database enlistment. Second, this study targeted the biomedical and health science journals published in Korea as a whole, but the distribution of journal characteristics and citation behaviors might be different in sub-areas of biomedical sciences [23]. Therefore, research in this area may be expanded by considering the specialty categories of biomedical and health sciences. Despite these limitations, the present study has value in that the relationship between database enlistment and citation rates was empirically analyzed to provide important information to enhance future journal circulation.

In conclusion, enlistment of a journal in prominent international databases is critical to international circulation. The results of the present study also showed that only low proportion of Korean journals in biomedical and health science fields were enlisted in prominent international databases. One strategy for enlistment is to publish journals in English, which would satisfy this prerequisite for enlistment in a prominent international database and help to attract more international readers. On the other hand, publishing domestic journals in Korean is advantageous for enhancing the circulation in Korea. Thus, the publication language should be carefully chosen by a journal with consideration of the target readers. In addition, more than 50% of Korean journals of biomedical science have not been enlisted even in KCI. To spread the content of biomedical journals published in Korea, more work should be done, such as indexing and bibliographic services initiated by reputable, governmental agencies, e.g., the National Center for Medical Information and Knowledge to provide validation and public trust.