Data illustrated here were retrieved from the Web of Science Core Collection (WOSCC), which is the most frequently used citation database for bibliometric analysis, and the structured data format provides considerable convenience to quantitative analysis. This study selects articles from four databases, namely, Science Citation Index Expanded, Social Sciences Citation Index, Conference Proceedings Citation Index-Science and Conference Proceedings Citation Index—Social Science & Humanities, as the data source.

Advanced search of WOSCC is used, and the research strategy was as follows: {[TS = (“integrat* care” OR “integrat* health” OR “integrat* healthcare” OR “care integration” OR “integrat* of care” OR “integrat* medic*”)]}. The publication time span was limited to 2010.10.1–2020.10.12, and 7,501 publications were collected. Document Type: Article.

Firstly, we use Excel’s bar graph to count the annual number of posts and the rate of change of posts (Figure 1). Secondly, we used the WOSCC Literature Analysis Report to find out the distribution of disciplinary areas (Figure 2). We then used BibExcel to extract the top 20 high-yield countries (Table 1; Persson et al., 2009). We used the CiteSpace visualization tool to depict the mapping of international productivity and collaboration (Figure 3; Chen, 2006). Next, we use R Voice’s BiblioMetrix to extract the number of articles posted by highly productive authors and other indexes (Table 2; Aria and Cuccurullo, 2017). Then, we used the Bibliographic Item Co-Occurrence Matrix Builder to extract the author co-occurrence matrix and imported SNA software Ucinet and NetDraw for drawing (Figure 4; Liu, 2009). We then used BibExcel to extract high-frequency keywords (Table 3) and used VOSviewer visualization software (Figure 5) to draw keyword co-occurrence knowledge mapping (keyword co-occurrence knowledge mapping) and the key in each word cluster (Table 4; van Eck and Waltman, 2010). Finally, we used CiteSpace to draw reference co-citation knowledge mapping (Table 5; Figure 6), highly cited documents (Table 6) and timelines (Figure 7). Specific tools and diagrams are shown in Figure 8.

Figure 1 presents the number of related papers published from 2010 to 2020. The number of publications on IC was rising in general. However, the annual growth rate of related literature varies greatly in this field; the highest literature volume is five times larger than the lowest (2012 and 2019).

Figure 2 shows the distribution of the disciplinary areas on IC. The total number of documents in the top 10 disciplines is 6,886, accounting for 91.107%, namely, health care science services (1,559, 20.627%), public environmental occupational health (1,166, 15.427%), health policy services (1,053, 13.932%), general and internal medicine (832, 11.008%), integrative complementary medicine (638, 8.441%), psychiatry (498, 6.589%), oncology (351, 4.644%), nursing (318, 4.207%), pharmacology pharmacy (259, 3.427%) and pediatrics (212, 2.805%).

Table 1 shows the top 20 countries ranked by the percentage of publications. The United States was the most productive country, with a total of 4,112, accounting for 54.82%, and the BC is 0.45, followed by the United Kingdom (656, 8.75%, 0.16) and Australia (501, 6.68%, 0.15). In terms of regions, Europe has 11 countries, accounting for 55%; Asia has 4 (20%), America has 3 (15%), and Oceania and Africa have 1 each (5% respectively). Combining Table 1 and Figure 3, the United States, with the largest number of publications and the highest BC, is at the center of the national cooperation network. In addition, countries such as the United Kingdom, Australia and the Netherlands have close cooperation with other countries, and they play important roles in the development of global IC. China has published 312 articles (4.16%) and ranks first in Asia, but the BC is only 0.08. showing that China lacks international cooperation with other countries on IC.

The BiblioMetrix Tool Package in R development software was used to analyze all the authors in 7,501 studies, with a total of 31,958 authors. The number of authors per paper was 4.26, and the number of authors with one paper was 25,411, accounting for 79.5%.

According to Price’s Law, authors of more than or equal to six articles (the core author), a total of 598 authors, and only the top 20 prolific authors are selected in this study. Then, software to calculate multiple indicators such as the number of essays, BC, CC, DC, h-index and g-index for the top 20 authors; multi-dimensional judgment of the author’s influences is developed (Hirsch, 2005; Egge, 2006). As shown in Table 2, Ben-Arye E has published the most articles, with a total of 60 articles, followed by Schiff E and Jacobsen SJ.

Figure 4 shows the collaborative network of the top 20 productive authors. The nodes represent 20 core authors; The circle means BC is 0, the diamond means BC is 2, the upward-pointing triangle means BC is 6, the box means BC is 7, and the downward-pointing triangle means that BC is 14. Combining Table 2 and Figure 4, although Quesenberry CP does not have the largest number of articles, the BC is the largest, so he is at the center of the network. In addition, Quesenberry CP, Jacobsen SJ, Sidney S, and Karter AJ formed an academic group, who are from the same institution (Kaiser Permanente, Oakland, California, United States). The other academic group was formed by Ben-Arye E, Schiff E, and Samuels N. However, no cooperation occurred between the two groups. In addition, six authors did not cooperate with other authors.

Noun items can be used to detect research hotspots in the field (Chiu and Ho, 2007). Keywords are nouns or phrases that reflect the core content of a publication (Chen et al., 2018), and high-frequency words represent the research hotspots (Xie et al., 2008). BibExcel was used to extract all the keywords in the literature, and after merging and removing duplication, a total of 12,449 keywords were obtained. Among them, 9,347 keywords appeared once, accounting for 75.08%. According to Price’s Law, keywords with a frequency greater than or equal to 24 are considered high-frequency keywords. The top ten keywords by frequency are “IC” (1005), “integrative medicine” (547), “primary care” (349), “mental health” (167), “complementary and alternative medicine” (137), “qualitative research” (131), “integration” (129), “complementary medicine” (115), “primary health care” (115), and “quality of life” (110), as shown in Table 3.

VOSviewer was used to draw map keyword co-occurrence networks, as shown in Figure 5.

A total of 4 clusters and 1,144 lines were formed based on the Linlog/Modularity algorithm. These 4 clusters represent the research hotspots on IC in the past decade. Using the largest node in each research hotspot as the cluster name, Cluster 1 (Red, IC) is the largest category and contains 36 keywords. Cluster 2 (Green, Depression) contains 26, Cluster 3 (Blue, Integrative medicine) contains 19, and Cluster 4 (Yellow, Primary health care) contains 12, as shown in Table 4.

The keywords in the four clusters reveal the characteristics involved in the development of integrated healthcare. ① IC should be value-based and health-centered, providing health services across the life cycle and achieving optimal health outcomes with limited resources. ② Service targets should include the whole population, not just special people who need long-term care.③ Focus more on health promotion and disease prevention, and to promote the integration of public health services and clinical care. ④ Pay attention to self-management and be the first responsible person for own health. ⑤ Attach importance to the role of complementary medicine in IC. ⑥ IC strategies need to be designed and developed taking account of the particular local realities.⑦ Promote incentive reform and give full play to the regulatory role of health insurance leverage to promote the development of IC. ⑧ Use digital technology to empower IC. ⑨ Establish effective governance, evaluation and accountability mechanisms.

Over the past 10 years, although the IC field has received increasing attention, the degree of attention has been gradually decreasing every year (Figure 1). This could be attributed to various factors, and one possible reason is that it has become more widespread, thus the initial novelty is fading and people are shifting their attention to other medical issues and priorities. While the attention given to IC may be decreasing, it is important to recognize its value and potential in improving healthcare outcomes. In terms of disciplinary area distribution (Figure 2), IC is widely distributed in multiple fields, such as health care sciences services, public environmental occupational health, health policy services, general and internal medicine, psychiatry, oncology, nursing, pharmacy and pediatrics, reflecting that the breadth and depth of IC applications are constantly expanding. The wide distribution of IC in various fields presents a unique opportunity for collaboration and knowledge exchange. For instance, healthcare professionals from different fields can share their insights and experiences in implementing IC in their respective fields, thus enhancing the development and delivery of IC services.

In terms of the number of publications and the BC value (Figure 3), the top three countries were the United States, the United Kingdom and Australia. Combined with Table 1, the proportion of developed countries’ publications is much higher than that of developing countries. Many high-income countries in Europe and America have adopted IC to address the health needs of aging populations and rising rates of chronic and multiple morbidities (Point, 2011; Kringos and Klazinga, 2014). For instance, IC has been available in the United States for 76 years, since Kaiser Permanente began offering the first IC insurance plan in 1945 (Craig et al., 1999; Laugesen and France, 2014). The English National Health Service has experimented with the integration of care since the 1990s and established 16 IC pilot programs with a range of objectives in 2008. To some extent, there is no one model of IC strategy that can fit all countries (Averill et al., 2010; Maruthappu et al., 2015), the design of specific measures should take into account the local culture, socioeconomic development, governance, and service recipients (Atun et al., 2010; Stadnick et al., 2019). However, even if different countries have different paths to achieve IC, there is a need to develop a uniform framework of principles. WHO has developed a common set of principles of IC that are comprehensive, equitable, sustainable, holistic, preventive, empowering, respectful, collaborative, co-produced, endowed with rights and responsibilities, governed through shared accountability, evidence-informed, led by whole-systems thinking and Ethical (World Health Organization, 2015). Furthermore, value-based healthcare has become an international consensus (Daniels et al., 2022; de Vasconcelos et al., 2022; Kokko, 2022). While WHO provides a thorough and widely-accepted IC framework, but there’s potential for further refinement, context-specific adaptation, and inclusion of new concepts as healthcare evolves. The principles’ effectiveness hinges on their practical implementation in real-world healthcare scenarios. Thus, countries should collaborate to enhance the value objectives, principles, and key elements of IC, and explore paths of integration tailored to each nation’s characteristics within the theoretical framework.

Bibliometrix R-tool was used to analyze the top 20 most productive authors (Table 2), Ucinet was used to manage the co-author matrix. Ben-Arye E had the most publications but a BC of 0 and was not placed in the control position in the network information; the author formed a triangular cooperation network with Schiff E and Jacobsen SJ. On the other hand, although Quesenberry CP only published 31 articles, he collaborated with several authors to form a large cooperative network. In addition, he had the largest BC and was at the center of the network, meaning that he acts as a bridge for information in the network. The analysis reveals that effective cooperation among authors is lacking, despite the formation of several scholar groups. This finding indicates a need for greater collaboration within the academic community, by promoting more effective collaboration, including theoretical exchanges and practical sharing, researchers can pool their expertise and resources to accelerate advancements in the field.

From BibExcel statistics, 93 high frequency keywords were extracted (Table 3). In addition to “integrated care” and “integrated health care,” the hot keywords included “primary care,” “mental health,” “complementary and alternative medicine,” “palliative care,” and “self-management.” By using the knowledge mapping tool VOSviewer to visualize keyword co-occurrence analysis (Figure 6), four further hotspot clusters were generated (Table 4). These results indicate that IC is not limited to the clinical field, but has broken down disciplinary barriers and encourages interdisciplinary cooperation, thus health systems should provide health services that integrate prevention, diagnosis, treatment, rehabilitation, and palliative care to cover the health needs of the region’s residents. In addition, to meet the needs and preferences of the population, individuals should be seen as active participants in health management and take responsibility for their own health (Rittenhouse and Shortell, 2009; Singer et al., 2011).

Science knowledge mapping was used to analyze reference co-citation, highly cited articles and timelines and detect and research historical evolution and trends. With the support of CiteSpace, 210,081 references were analyzed. According to document co-citation analysis, seven clusters were generated (Figure 6), namely, Cluster 0 (qualitative study), Cluster 1 (serious mental illness), Cluster 2 (health systems integration), Cluster 3 (integrative medicine), Cluster 4 (patient-centered medical home), Cluster 5 (chronic pain) and Cluster 6 (complementary medicine). We observed that the prominence of Cluster 0 (qualitative study) as the main cluster in keyword analysis could be due to its vital role in integrated healthcare research, where it is often used to investigate patient experiences, healthcare behaviors, and healthcare decision-making. Furthermore, we noted that the earliest research clusters are Cluster 3 (integrative medicine) and Cluster 4 (patient-centered medical home), which are related to the history of integrated healthcare development, from advocacy IHCD to encourage health systems to embrace PCIC approach to organizing health services. Notably, “complementary medicine” has become a hotspot in recent years, especially with the increasing prevalence of cancer, chronic diseases and mental illnesses; integration of complementary and western medicine can improve the quality of life of patients (Mongiovi et al., 2016; West, 2018; Saeed et al., 2019; Hübner, 2020). In addition, traditional Chinese medicine plays a positive role in the treatment of COVID-19 (Ren, 2020; Zhang et al., 2020). Thus, complementary medicine will continue to be a focus in the future, however, it should be noted that the use of CAMs in clinical practice needs to assess the effectiveness, and safety of CAM (Paoloni et al., 2022).

Among the top 10 frequently cited articles, four were concerned with integrated or integrative care, and the other articles were about pediatrics, collaborative care, complementary and alternative medicine, highlighting the interdisciplinary nature of IC. These highly cited articles were contributed by influential authors such as Valentijn PP, Koebnick C, Berwick DM, Barnett K and Woltmann E, who have made significant contributions to the development and advancement of the field, enhancing our knowledge and understanding of IC. The sources of articles with more citations and higher influential factors included the Lancet, the New England Journal of Medicine, the International Journal of IC and the American Journal of Psychiatry. High-quality periodicals have a positive effect on the development of the subject. These journals provide a platform for disseminating research findings and facilitating collaboration among researchers and healthcare professionals, contributing positively to the development of the field.

The advancement of IC faces different challenges, especially in complex and changing external environments. Firstly, theory and practice are not in sync. Person-centered care is often defined as the core of IC, so people-centered care needs people-centered research, IC strategies should be based on co-creation with patients or citizens in ways that involve, engage and empower them, but a chasm remains between theory and practice (van der Vlegel-Brouwer et al., 2020). In addition, inter-organizational collaboration is the key to delivering IC. However, many barriers related to administration and regulation, resources and funding hinder the collaboration (Andersson et al., 2011; Koebnick et al., 2012; Boothroyd et al., 2015; Auschra, 2018; Simpson et al., 2023), Therefore, relevant authorities should take action to overcome such barriers. For example, China has proposed the “County Medical Alliance,” which integrates the medical service network at the county, township and village levels in six areas: administration, personnel, finance, services, assessment and supply of medicine and equipment to provide a full, continuous and coordinated service for service recipients. Countries or regions with health service systems similar to those of China can learn from the “County Medical Alliance” model to promote the development of IC. Notably, financial incentives are potentially powerful tools to stimulate IC (Averill et al., 2010; Tsiachristas, 2016), but separate payment mechanism can block effective integration (Struckmann et al., 2017), moreover, each payment method has a certain negative incentive effect, so mixed payment should be adopted to support more effective and efficient IC system (Berenson and Rice, 2015; Stokes et al., 2018). Furthermore, silos of data collection on a cross-institutional level have created a fragmentation of electronic medical records in many countries, potentially hindering continuity of IC and resulting in clinical, and administrative inefficiencies (Bradley et al., 2017; Meinert et al., 2019; Eh et al., 2020). Governments should fully recognize the great value of medical data, strengthen information systems and promote legal and compliant medical data sharing.