Trend of Women's Health Research in Korea, 2012-2020: Topic and Text Network Analysis.

Women’s health research is the study of health across a woman’s life course, including all health conditions for which women and men experience differences in risk, presentation, and treatment response, as well as health issues specific to women. 1 The field has expanded beyond reproductive health and covers the study of health throughout the lifespan and across the spectrum from basic research to clinical trials. 2 Recently, several countries, including the United Kingdom, 3 4 the United States, 5 and Canada 6 established national strategies to broaden and deepen the landscape of women’s health research. As women have historically been underrepresented in clinical studies and trials, 7 8 9 further evidence is required to fully understand women’s diseases or conditions throughout their life course. Globally, research for women’s health had primarily focused on the anatomical and biological differences between men and women, 10 and yielded significant progress in the treatment of breast cancer, cardiovascular disease, and cervical cancer. 11 The recent national strategies for women’s health research now aim to encompass the diverse health needs of women, along with evidence-based disease prevention and treatment tailored to women’s specific needs. 11 Limited knowledge about sex- or gender-based differences and sex-specific health conditions adversely affects women’s health care. Without updated information on the topics and contents of women’s health research, research funders, policymakers, and researchers lack a basis for allocating resources required to address the problems of women’s health. In prior analyses in 2008 and in 2012, the majority of studies on women's health have concentrated on breast and cervical cancers in Korea. 12 13 Studies on health promotion, risk factors of disease, chronic diseases, or conditions relevant to women’s health are relatively scarce. It has been recommended that research topics in women’s health address a broader range of issues that affect Korean women’s lives. 12 There are increasing demands for new research regarding women’s health issues in Korea. These issues, raised by civic or advocacy groups as well as established researchers, include research topics related to recent changes in social institutions and culture. Examples of new research demands may include safety issues regarding sanitary products, beauty products, or cosmetic procedures or surgery and their impact on health 14 15 16 17 ; studies on the vulnerable subgroup among women, such as the disabled, migrants, sexual minorities, vulnerable workers, the elderly 18 19 ; and research on the physical and mental burden on the role of caregiver, which is mainly filled by women. 20 21 There is also new research interest regarding the impact of novel technology and its evolution on women’s health. 22 Given these diverse research demands, it is necessary to analyze the recent status of women’s health research in order to determine future research directions. The present study aimed to examine the trend of women’s health research and to analyze research topics supported by the national funding program in 2012–2020. Through text mining and content analysis, we explored the national initiative for women's health research in Korea and discussed its limitations and possibilities.

In this study, the term “female” was used to indicate one anatomical sex according to the biomedical definition, and the term “women” referred to gender identity shaped by the individual’s experience, presentation, and environment. 9 To identify women’s health research projects funded by the government, we searched all final reports of research projects published during 2012-2020 using the National Science & Technology Information Service database of Korea (NTIS, https://www.ntis.go.kr/ThMain.do ) in July 2022. Because there is no single search indicator for women’s health research, the search terms included a core keyword such as “women” or “female” combined with other search terms, based on the US Institute of Medicine (IOM)’ s classification 11 as it provides a framework of concept and represents the characteristics of women’s health research. These five categories are: 1) health promotion-related, 2) disease-related, 3) organ-related, 4) reproduction-related, and 5) population-related. Search terms in categories 1) and 2) included general and common terms in health research and terms representing diseases or conditions that are common to or serious for women ( Supplementary Table 1 ). Search terms in categories 3), 4), and 5) included terms principally designating female-specific organs and health conditions, and women-dominated populations. The search terms in these categories were used individually in the search engine. After obtaining the entire list of research projects, information on the study ID, the launching year, title, abstract, funding, researcher affiliation, and governmental source of funding were extracted. If the research project was a multi-year task, its annual research reports were considered individual research projects. The initially extracted database contained 6,175 research projects, of which 5,624 projects remained after excluding the duplication list ( Fig. 1 ). All 5,624 research reports were manually screened to determine whether they were relevant to women’s health research, and were excluded if their research topic was: 1) Not a research topic about women’s health or healthcare field 2) About basic biology that is not directly related to women's health and diseases 3) Mainly about fetuses or infants, not maternal health 4) About issues related to the development of commercial female products, which are not classified to medical supplies. NTIS = National Science & Technology Information Service database of Korea. Two researchers (Park J and Min HS) independently examined and cross-checked the report’s title and abstract until a consensus was reached on the final list of projects (n = 3,269) after several rounds of discussion ( Fig. 1 ). A descriptive analysis was undertaken for research trends by year and source of funding. To examine the scope and extent of research topics on women’s health, both deductive and inductive approaches to content analysis were performed. Initially, researchers used three topic categories based on the US IOM’s classification 11 which included diseases or conditions that are common to or more serious in women (category 1), diseases or conditions that are specific to women (category 2), and factors that are known as determinants of health (category 3; Table 1 ). In the process of topic classification using this framework, the researchers inductively generated subcategories through constant classification and comparison, and iterated topic classification according to the subcategories. To determine the reliability of the whole coding framework, the first researcher classified all 3,269 topics, and the second researcher randomly classified 10% of the lists. This comparison showed very good inter-coder agreement by Cohen’s kappa (> 0.9). We stratified the projects according to the six funding entities considering changes in the administrative division of the governmental ministries and funding size (Health and Welfare, Food & Drug; Science, Technology, Education; Trade, Industry and Energy; Environment, Labor, Culture, Gender Equality and Family; Policy and Coordination of Prime Minister; Multi-department). Those funded by the Ministry of Science and ICT (MSIT) and the Ministry of Education (MOE) were regarded as having the same funding source because both were under the Ministry of Education, Science, and Technology (MEST) in 2008–2013. Similarly, the Ministry of Health and Welfare (MOHW) and the Ministry of Food and Drug Safety (MFDS) were regarded as the same funding source, as the MFDS was under the MOHW until 2013. Keywords with a minimum of 10 occurrences from the title of the research project were extracted for network visualization using VOSviewer version 1.6.18, a software tool for constructing and visualizing bibliometric networks (Centre for Science and Technology Studies, Leiden University, Leiden, Netherlands), which clustered keywords for topic analysis. In the figures generated by VOSviewer, the size of keyword circles represents the keyword’s frequency of occurrence and co-occurrence links with other keywords, and the distance between circles reflects how frequently they co-occur. 23 Colors of clusters represents the topic clustered by keywords with co-occurrence links. For a more in-depth analysis of the research topics, text network analysis was conducted using NetMiner software (NetMiner Social, Seongnam, Korea), a network analytic tool for exploratory analysis and visualization of large semantic data. Keywords were extracted from the abstracts of research reports in 2012 (n = 227) and 2020 (n = 571), excluding stop words, and the occurrence frequency and degree centrality for keywords were calculated. As degree centrality measures the total number of keywords linked to a particular keyword representing frequency of co-occurrence among keywords, it can exhibit a “centrality” of a particular keyword in a keyword network. 24 Ethical review and approval were not required for this study because publicly available data from the government website were used.

The total number of research projects on women’s health was 235 in 2012, and 571 in 2020 ( Table 2 ). Between 2012 and 2020, the number of women’s health research projects increased by 2.4 times (12.9% per year on average) ( Fig. 2 ). The total research funding increased by 2.2 times (11.7% per year on average), from 25.8 billion Won in 2012 to 57.7 billion Won in 2020 ( Table 3 ). As the total research funding in the healthcare sector according to the Korean standard classification of science and technology was 1,104 billion Won in 2012 and 2,099 billion Won in 2020 25 26 ; the funding of women's health research accounted for 2.3% and 2.7% of the total research funding in the healthcare sector in 2012 and 2020, respectively. All values are presented in million Korean Won. When the number of projects was stratified by funding entities, the MSIT and the MOE supported most of the women’s health projects (2,334 projects comprising 71.4%) between 2012 and 2020, followed by the MOHW and the MFDS (682 projects, 20.9%) ( Supplementary Table 2 ). The MOHW, the MFDS, and the Ministry of Trade, Industry and Energy (MOTIE) accounted for a higher percentage of funding (27.1% and 20.0%, respectively) compared to the percentage of project numbers (20.9% and 5.0%, respectively) ( Fig. 2 , Supplementary Table 3 ). In category 1, where the diseases or conditions that are more common in women are included, breast cancer accounted for the highest percentage in the number of projects (40.7%) and in the funding (42.2%) over the period 2012-2020 ( Tables 2 and 3 ). In category 2, which covers diseases or conditions specific to women, research projects on gynecological cancers and sexual and reproductive health (19.1% and 16.7% for the number of projects, respectively) dominated the five subcategories. From 2012 to 2020, the number of projects showed the largest increase in the subcategory of other health determinants (8.8 times), followed by sexual and reproductive health (6.1 times), nonmalignant gynecological disorders (3.5 times), and breast cancer (2.2 times) ( Fig. 2 , Table 2 ). The change in the volume of funding was different across subcategories, with surging growth in sexual and reproductive health (11.1 times; Table 3 ). Proportions of researcher’s affiliation according to the research topic was added in Supplementary Table 4 . Two keyword clusters were identified by color in 2012 ( Fig. 3A ), mainly focusing on ‘breast cancer’ and ‘development.’ In 2020, keyword clusters outlined a more complicated linking structure ( Fig. 3B ). Five clusters were identified by color, of which main keywords included ‘women,’ ‘pregnancy,’ and ‘health’ (green), ‘breast cancer,’ ‘diagnosis,’ ‘gene,’ and ‘cervical cancer’ (red), ‘regulation,’ ‘pathway,’ and ‘drug resistance’ (yellow), ‘stem cell,’ ‘efficacy,’ and ‘menopause’ (blue). Regarding the government ministries that supported research funds ( Fig. 4 ), the MOHW and the MFDS funded research projects of which keywords were clustered to ‘women’ and ‘analysis’ (red), ‘pregnancy’ (blue), ‘clinical trial’ (purple), ‘metastasis’ (yellow), ‘preeclampsia’ (green). Other ministries funded various projects with focus on ‘cancer cell’ (green), ‘elderly women’ (red), and ‘metastasis’ or ‘stem cell’ (blue), alongside a focus on ‘postmenopausal women’ (pink), ‘experience’ and ‘gender’ (brown). MOHW = Ministry of Health and Welfare, MFDS = Ministry of Food and Drug Safety. To better understand the women’s health research topic in detail, the abstract of the research project was analyzed by text mining. When listing 30 frequently occurring keywords and their degree centrality index, keywords including ‘cell,’ ‘breast cancer,’ ‘patient,’ ‘treatment,’ and ‘tissue’ appeared most frequently both in 2012 and 2020 ( Supplementary Table 5 ). Keywords such as ‘endometrium,’ ‘uterus,’ ‘pregnancy,’ ‘imaging,’ ‘biomarker,’ ‘health,’ and ‘data’ appeared frequently in 2020, while ‘survey,’ ‘pathway,’ and ‘uterine cervix cancer’ were frequent in 2012. Overall, the degree centrality index increased proportionately as the occurrence frequency of the keywords increased. The degree centrality indices were relatively higher in 2020 than in 2012, which meant that some research topics received more attention from researchers.

Both the number and funding amount of women’s health research expanded between 2012 and 2020 in Korea. Despite its quantitative growth, the scope of research on women’s health in Korea has not significantly changed during that same period, except for the striking growth of sexual and reproductive health research. Most of the women’s health research has been on a limited range of research topics, especially cancer (including subcategories of breast and gynecological cancers) and sexual and reproductive health. In particular, sexual and reproductive health showed a steep increase in the number of projects and funding size, which might be promoted as part of the national initiative to reverse the trend of ever-decreasing fertility rates in Korea. 27 In line with these results, the keyword network of 2020 showed a more complex pattern than that of 2012, while still focusing mainly on cancer and reproductive health. Using topic and text analysis, we present the linkage of research keywords of publicly funded women’s health research in Korea. Overall, women’s health research has roughly doubled in terms of the number of projects and funding, but its share in the total health funding has increased by only 17% from 2012 to 2020. This would be attributed to that absolute amount of women’s health research funding was very small compared to the total healthcare investment during the period. For example, the total budget for the Korea Disease Control and Prevention Agency for ‘Women’s Health Technology Development and Research Project,’ the only targeted research project related to women’s health, was slightly above 1.5 billion won in sum in 2021. 28 Although the National Assembly audit for the MOHW’s R&D have called for the activation of women’s health research and multidisciplinary research across the life cycle since 2010, it is unlikely that national interest in women's health research has increased for the last decade. 29 In the current result, the majority of women’s health research projects in Korea during the last decade were similar to those of the prior period in terms of the topics. 12 13 We found fewer studies on chronic diseases and conditions, mental health, adolescent health prevention and risk factors, and policies and institutions, although these subjects embrace issues having a significant impact on morbidity and mortality for Korean women. 13 The dominance of breast cancer research in the field of women’s health research appears to be attributable to several factors: public awareness of its higher incidence, the development of innovative anti-cancer drugs, a well-established research network, diverse funding resources, good accessibility to biomaterials, and the active participation of patients in clinical trials. 11 30 31 32 33 Given the high burden of musculoskeletal disorders, diabetes, urogenital disorders, and blood and endocrine diseases in Korean women, more attention is required to diverse health issues across the life course of women. 34 The lack of diversity in women’s health studies was also illustrated shown by the current result of text analysis. The networks of research keywords in both 2012 and 2020, respectively, showed a higher occurring frequency and the centrality index of “cell,” “breast cancer,” “patient,” and “development.” The importance of those four keywords has even grown in 2020, showing a much higher occurring frequency and centrality index than other keywords. However, more various keywords were emerged in 2020 than 2012; there were only two keywords with occurring frequency of 100 or more (“cell,” “breast cancer”) in 2012, whereas it reached 15 keywords in 2020. Overall, the text network analysis revealed a continued dominance of breast cancer research, despite the expansion of research on other cancers and sexual or reproductive health, and an increase in the diversity of subtopics. Given the area of women’s health research covers gender difference and health determinants, our results highlight the paucity of health research considering sex and gender specificity in common diseases, as well as women-specific diseases or conditions. Taking sex and gender specificity into account is important for improving diversity in women’s health research. For example, clinical and epidemiological manifestations an d responses to treatment of common cancers would differ between women and men depending on health behaviors, dietary risk factors, and the sensitivity of the reproductive system. 35 36 37 In its commitment to the Gender Equality Plan, the European Commission (EC) requires all research and innovation projects funded by the EC to integrate the gender dimension. 38 Since 2016, the National Institutes of Health (NIH) of the US has asked grant applicants to explain how sex and gender are factored in research designs and analyses for studies in vertebrate animals and humans. 39 In reference to these cases, we suggest adding an item about the contribution of the research to the knowledge of sex or gender differences in the application for health research grants by the government, as the same preventive measures, diagnostic, and treatment protocols may have different effects or be utilized differently depending on sex. It is noteworthy that sexual and reproductive health showed a steep increase in the number of projects and funding size, which suggests the possibility that public investment in this research field to increase the fertility rate have been carried out as the low birth rate in Korea has deepened. The MOHW have established the Health and Medical Technology Fostering Basic Plan every five years since 2012, and it included the development of responding technologies to low birth rates and population aging as a detailed task. 40 Accordingly, the MOHW has invested research funds in the development of fertility preservation technology, the research for miscarriage and premature birth, or fertility-reducing disease treatment and technology. 41 As of October 2017, furthermore, artificial reproductive technologies began to be covered by the National Health Insurance, which might greatly facilitate research on fertility treatments by increasing patient participation. 42 Similar to the research subjects of funding from MOHW, pregnancy-related topics predominate in the research field of sexual and reproductive health in our study. The visualization of keyword networks showed that research topics were mainly focused on issues related to pregnancy, centering on keywords such as preeclampsia, preterm delivery, placenta, or prenatal diagnosis. However, providing comprehensive health services that cover not only the health of women during pregnancy but also the prevention of pregnancy or preparation for pregnancy is essential. 43 World Health Organization recommends a more comprehensive approach to research on sexual and reproductive health, which requires an evidence-based, respectful, and positive approach to sexuality and sexual relations free from all types of coercion and violence. 44 Reproductive health can include a wider range of issues, from the physiology of puberty, postnatal rehabilitation, cultural meanings, and vulnerable populations to unwanted pregnancies, abortion in safe conditions, and gender inequities in the local context. 45 46 Beyond the issue of low birth rate and pregnancy, research on women’s health in Korea should broaden its scope to utilize a comprehensive approach in sexual and reproductive health, based on women’s needs. 47 48 To examine research topics conducted without funding in Korea may provide insight to understand the scope of research topics that have requested public research funds. According to the prior results of analyzing women’s health studies in major nursing journals in Korea, in which unfunded studies account for about 70%, topics focused to motherhood research, menstrual issues, sexuality or sexual life, plastic surgery and violence, and other social influence on health. 49 Although it was not disclosed whether studies were funded or not, 302 studies searched using ‘women’s health’ from 2010 to 2020, in Korea, had research keywords including marriage migrant woman, mental health, multicultural, female worker, etc. The authors argued that it is necessary to expand the scope of research issues to various topics by reflecting social trends and changes. 50 To cover such diverse topics and expand the scope of topics in women's health research, the role of government agencies as a major funder is pivotal. 33 Despite uncontrollable factors influencing research priorities and funding, public research institutions can maintain awareness of the uniqueness of women’s health and the disease burden. In the US, the Office of Research on Women’s Health (ORWH) provides funds through NIH to augment new research initiatives, and adequately address high-priority issues concerning women’s health. 49 51 52 To pursue the research priority required by women of all groups, the government agency should be able to converge the research demands from diverse interest groups and find topics that are under-studied in the field. 33 53 54 55 56 57 This study has limitations that require the findings to be interpreted with caution. First, although the search terms for women’s health research were selected in a structured manner, the search results might not be the complete list of all women’s health research. However, considering a search strategy based on the criteria of the US IOM, we believe that our analyses covered almost all eligible projects on women’s health research nationwide. Second, the classification of the research topics was based on the title and abstract, which may lead to misclassifications when they do not fully reflect the content of the project. However, given that project researchers are required to present the methodology and key findings in their abstracts, the major characteristics of the studies should be properly reflected. Third, we did not analyze the detailed process of the review and allocation of funding by governmental entities. With more detailed information on the review and decision process, we will be able to identify the factors that have influenced the scope of women’s health research in Korea. In summary, the topics of women’s health research in Korea need to cover more diverse health challenges faced by women in varying stages of their lives. To better understand women’s health needs and include both gender and public health perspectives, the implementation of corresponding guidance in funding policies is required. Several policies can be proposed to advance women’s health research in Korea. First, the public research program should focus on topics that are the main causes of morbidity and mortality for Korean women but remain under-researched. Second, a government research agency dedicated to women's health should enhance research by setting research priorities and principles, institutionalizing communication channels with various interest groups, and building a research network with a long-term perspective.