Research status and trends analysis of long-term care: A bibliometric analysis

Abstract

Long-term care (LTC) is a fundamental system in many countries that provides essential aged care services and promotes the physical and mental well-being of older adults. This study visually maps the progress and trends in LTC research, offering theoretical references for fellow scholars and practitioners. Using the Web of Science Core Collection as the data source, we retrieved publications on LTC from 2010 to 2024. CiteSpace bibliometric software was employed with a time slice of 1 year and a selection threshold of top 50 per slice. Co-occurrence and collaboration networks were analyzed for keywords, authors, and institutions. The United States contributed the highest number of publications (1896). Among journals, The Lancet published the most articles (2568). The University of Toronto exhibited the strongest centrality (0.23). The top 5 most frequent keywords were “long-term care” (1666 occurrences), “dementia” (726), “nursing home” (674), “health” (615), and “older adult” (551). Qualitative synthesis further identified major research themes including long-term mortality among LTC recipients, nursing home residents, LTC insurance, pragmatic trials, and social isolation. Author collaboration networks showed only small clusters with weak overall connectivity; institutional collaborations were also limited, predominantly involving universities. This fragmented cooperation pattern did not improve significantly over time, providing a critical benchmark for evaluating the effectiveness of academic community building within the field. Notably, all high-frequency keywords exhibited extremely low centrality values (0.01–0.02), indicating that while a large volume of research centers on a few core terms, the knowledge linkages between these topics and broader research issues remain weak. Additionally, burst detection revealed strong emergence of terms such as “unit” and “setting,” further highlighting an exceptionally high concentration of research attention on micro-level care facility issues. Future research should strengthen international collaboration at the macro-level, expand the target population of LTC to a wider range of older adults at the meso level, and address the chronic disease needs of older individuals at the micro-level, thereby advancing the development of LTC systems.

1. Introduction

Population aging and services for the elderly are issues garnering attention from governments, society, and academia all over the world. World Population Prospects 2022: summary of results shows that the share of the global population aged 65 and over will grow from 9.7% in 2022 to 16.4% in 2050, and that the global average life expectancy will increase from 72.8 years in 2019 to 77.2 years in 2050.^[1] The sharp increase in the number of elderly people and the extension of life expectancy will lead to a rising proportion of disabled and semi-disabled elderly people and the increasing prevalence of chronic diseases among the elderly,^[2] which have attracted great attention from governments around the world. These governments have proposed to provide continuously improved long-term care (LTC) services for the elderly. According to the Organization for Economic Cooperation and Development, LTC refers to the provision of a range of services for people in need of help with basic activities of daily living over a long period of time.^[3,4] LTC is a crucial strategy for countries to cope with the aging of their populations. Europe recognized early on that the LTC needs of the elderly are not only the responsibility of the family but also require the cooperation of the entire society.^[5] Therefore, some countries have begun to design and implement LTC policies suited to their national conditions and to establish and improve comprehensive and continuous support for the elderly in the form of health education, preventive insurance, disease treatment, rehabilitative care, hospice care, and pension funds. Based on different theoretical systems, scholars have conducted a great deal of research on the connotation, model, influencing factors, policies, dilemmas, and countermeasures of LTC from different perspectives and by adopting different research methods.^[6–9]

This paper conducts a review of prior scholarly research, aiming first to synthesize existing reviews of the LTC literature. Li Pingfu et al conducted a comprehensive scientometric review based on 14,019 LTC literatures retrieved from the Web of Science Core Collection databases from 1963 to 2018 to explore the current status and trends of LTC research worldwide.^[10] Sun Zhaohui et al found that the psychological and physical health status of older adults in need of LTC after the COVID-19 outbreak shifted to the impact of the pandemic on older adults in LTC facilities.^[11] Pot et al conducted a literature review on the supervision of LTC.^[12] Wang et al assessed the quality and safety of care in a LTC environment using COVIDENCE for screening, data extraction, and quality assessment.^[13] Scholars have obtained a large number of theoretical and empirical results and have continuously expanded the theory of LTC. Overall, there are more qualitative studies in the LTC research review, but relatively few quantitative studies, and those are relatively outdated. Therefore, this study employs CiteSpace – a widely utilized information visualization software for bibliometric analysis and mapping of scientific knowledge – to analyze LTC literature from the core journals indexed in the Web of Science database. Through systematic literature review and analytical mapping, CiteSpace enables the visualization of knowledge structures, identification of research hotspots and frontiers, and tracing of evolutionary trajectories. This approach allows researchers to move beyond the limitations of traditional reading and narrative synthesis, offering a more systematic, objective, and intuitive insight into the holistic development of an academic field. This paper focuses on LTC literature published between 2010 and 2024. In addition, it incorporates a qualitative review of literature from the past 5 years. This integrated methodology addresses 2 common shortcomings in prior research: the reliance on outdated datasets in earlier CiteSpace-based bibliometric analyses, and the conventional separation between quantitative and qualitative approaches in LTC studies. The aim is to comprehensively understand the current state, intellectual evolution, and emerging trends in LTC research, thereby providing directional recommendations for further theoretical inquiry and practical references for the ongoing development of LTC systems.

2. Research methods and data sources

2.1. Research methods

The main method used in this study is bibliometric analysis, and CiteSpace was utilized as a literature analysis tool. CiteSpace develops software to explore key nodes and knowledge evolution paths in the literature based on Popper’s Three Worlds Theory, Price’s Theory of Scientific Frontiers, Bott’s Structural Hole Theory, and Information Foraging Theory.^[14,15] It can explore the internal relations and rules of the literature and show them in a visual way. In this study, “author” and “keyword” with high frequency were selected as network node types to generate co-occurrence maps for visual analysis. Based on this, combined with traditional literature review methods, first, employ CiteSpace software to conduct a network property analysis of the centrality of collaborating institutions and authors, generating institutional collaboration network maps and author collaboration network maps along with their centrality metrics. Second, perform a keyword cluster analysis using CiteSpace to generate keyword co-occurrence cluster maps and timeline views. Finally, utilize the citation burst detection feature in CiteSpace to analyze the varying activity levels of keywords across different time periods. Additionally, on the basis of the bibliometric analysis, integrate traditional qualitative literature review methods. Hot research areas of LTC were analyzed and its future evolution trends were explored.

2.2. Data sources

The data for this study were sourced from the Web of Science (WoS) Core Collection, which includes its constituent databases such as SCI-EXPANDED, SSCI, A&HCI, CPCI-S, CPCI-SSH, CCR-EXPANDED, and IC. A comprehensive search was conducted to retrieve literature on LTC published between January 1, 2010, and June 30, 2024. The WoS Core Collection was selected as the data source given its extensive coverage across multiple research disciplines, ensuring that the analysis is based on literature from internationally authoritative journals. The search criteria for the WoS Core Collection database were “Database selection Web of Science Core Collection” AND “Title = Long term care” AND “Publication date = 2010-01-01 to 2024-06-30” AND “Literature type = Essays and review papers” AND “Language = English.” It is particularly worth noting that this study employed the “Title” field for the literature search to screen for publications closely related to LTC. The primary reason is that a search using the “Topic” field (which covers titles, abstracts, and keywords) would have yielded 259,830 records. Given the excessively large volume, manually removing literature irrelevant to LTC would have been highly challenging. This approach would also have risked including numerous unrelated publications, potentially compromising the accuracy of the study’s findings. Therefore, the “Title” field was selected for the search, resulting in the retrieval of 10,769 publications. The retrieved literature was imported into the NoteExpress software. An initial screening was performed using the software’s filtering functions, followed by a meticulous manual review to exclude records that did not meet the inclusion criteria. The exclusion criteria were as follows: (1) records with missing critical information, such as publication year, author name, affiliated institution, or keywords; and (2) irrelevant publication types, including book reviews, forum announcements, and editorials. After removing 3756 duplicate records, a final corpus of 7013 articles was established (Fig. 1). This refined dataset was then organized and exported in plain text format, serving as the foundational data for subsequent analysis. The study explored this corpus from multiple dimensions. This study is a systematic review conducted in strict accordance with the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 statement to ensure transparency, completeness, and reproducibility of the research process.^[16]

Figure 1.

PRISMA flow diagram. PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

2.3. Data processing

Using CiteSpace.v.5.7.R5 was used to process and analyze data, and the time span of the WoS dataset is from January 1, 2010 to June 30, 2024. The time slice was set to 1 year, and the selection criteria were set to “Top 50” (Table 1).^[17]

Table 1.

Parameter settings.

Parameters
Time slicing	Monthly (y = 0) and yearly (y = 1)
Pruning	Pathfinder and pruning sliced networks
Node type	Country, journal, institution, author, keyword
Links	Strength (cosine) and scope (within slices)
Selection criteria	G-index (k = 50)
Visualization	Network analysis, term frequency analysis, co-occurrence and cluster analysis, burst detection analysis

3. LTC research combing

3.1. Analysis of the distribution of research countries

From the WoS-published LTC research papers, the United States (USA) published the most papers, as high as 1896 papers, accounting for 27.03% of the total statistical analysis of the literature, followed by Canada (CANADA), with a total of 930 papers, accounting for 13.26% of the total. The top 10 countries in terms of the number of papers are the United States, Canada, the United Kingdom, Japan, Germany, the Netherlands, Taiwan of China, Mainland China, Australia, and South Korea. In terms of mediator centrality, the mediator centrality analyzed by CiteSpace statistics can evaluate the importance of the node in the network. The mediator centrality of Canada, the United States, Australia, the United Kingdom, and Brazil are 1.21, 0.38, 0.36, 0.16, and 0.13, respectively. It can be seen that Canada, the United States, and Australia are leading in the number and importance of studies in the field of LTC research. While Canada’s research output in LTC is only about half that of the United States, its centrality in the collaboration network is remarkably high, far exceeding that of other countries. This indicates that Canada acts as the most critical “hub” or connector within the global collaborative network for LTC research. Conversely, the United States, despite having the highest volume of publications, exhibits a relatively lower centrality, which is not proportionally reflected in its output dominance.

3.2. Analysis of the distribution of articles according to time series

From the time of WoS publication, the number of publications has increased incrementally year by year, with 265 published in 2010, 340 published in 2015, growing to 676 in 2020, reaching a peak in 2022 with a total of 833 published. In 2023, the number slightly declined to 722, and from January to June published a total of 249 articles were published. Obviously, there is an upward trend in overall issuance, with a decline in 2023 after a gradual increase in research in the field of LTC, indicating that research in LTC has reached a mature stage, and the specifics of WoS issuance are shown in Figure 2.

Figure 2.

WoS-published papers in the field of long-term care research from 2010 to 2023.

3.3. Distribution analysis of journals

From the perspective of journals that publish LTC research, Table 2 shows that influential journals such as The Lancet, New England Journal of Medicine, and Journal of the American Medical Association (JAMA) include many academic papers on LTC, which to a certain extent indicates that there is a great deal of international attention to this issue. There are also many journals in the field of medical and health policy that include papers on LTC, which means that the issues of LTC have been emphasized by journals in the field of medical and health policy.

Table 2.

The top 10 journals by the number of articles published in the combination of long-term care from 2010 to 2022.

Wed of Science Core Collection
Serial number	Journal name	Number of articles	Impact factor	Serial number	Journal name	Number of articles	Impact factor
1	Journal of the American medical directors association	2568	4.669	6	Bmc health services research	876	2.655
2	Plos one	1891	3.24	7	BMC Geriatrics	615	3.921
3	Bmj open	1618	2.692	8	BMC public health	582	3.295
4	Cochrane database of systematic reviews	1479	9.266	9	Trials	505	2.279
5	International journal of environmental research and public health	992	3.39	10	Journal of the American geriatrics society	473	5.562

3.4. Analysis of cooperating institutions and authors

Through CiteSpace’s scientific analysis of scholars, institutions, countries or regions, etc, it is possible to find out the situation of authors, institutions, countries or regions issuing articles and cooperation and the connecting lines between the nodes of the authors, institutions, countries or regions reflect their cooperation, and the more connecting lines between the nodes indicate the stronger cooperation intensity.^[18,19]

3.4.1. Analysis of partner institutions

In terms of WoS cooperative institutions, there are 569 network nodes, 2597 node connections, and a network density of 0.0161. The research collaboration in the field of LTC is relatively loose on a global scale. The top 5 institutions in terms of global publications are the University of Toronto (centrality 0.23), McMaster University (centrality 0.12), University of Waterloo (centrality 0.06), University of British Columbia (centrality 0.06), and the University of Alberta (centrality 0.05) (Fig. 3). The relatively low network density indicates that while international collaboration in this field is geographically widespread, the strength of these connections is limited, resulting in an overall loosely structured global collaborative network. Furthermore, a significant number of the prolific publishing institutions are based in Canada, which aligns with and provides institutional-level evidence for the earlier finding that Canada exhibits the highest centrality. The University of Toronto demonstrates a centrality far exceeding that of other institutions, signifying its role as the most critical “bridge” or hub within the global collaborative network for LTC research.

Figure 3.

The cooperation of institutions in the field of long-term care research.

3.4.2. Analysis of authors’ cooperation

In terms of author cooperation in the WoS core collection, there are 795 network nodes, 1869 connections, and a network density of 0.0059 (Fig. 4), indicating less author cooperation in the field of LTC research. In terms of author publications, the top 5 authors are Sharon Kaasalainen (51 publications, centrality 0.08), Katherine S. Mcgiltion (30 publications, centrality 0.01), Carole A. Estabrooks (25 publications, centrality 0), Thomas Hadjistavropoulos (23 papers, centrality 0), John P. Hirdes (23 papers, centrality 0.01), and Nanako Tamiya (21 papers, centrality 0). The author collaboration network exhibits a very low density, indicating that within a large network comprising 795 authors, the actual collaborative connections (1869 links) are relatively sparse. This suggests that collaboration among leading authors in LTC research is generally broad in scope but weak in strength, with most researchers likely confined to fixed, small-scale teams, and extensive cross-team or cross-institutional collaboration has yet to become the norm. The most prolific author in the network demonstrates a centrality of only 0.08, while the top 5 authors by publication output show centralities of zero or near 0. This finding indicates that even the most productive scholars do not play a central role in bridging different research groups or facilitating knowledge flow across the network. When combined, the low network density and the low centrality of high-output authors suggest that the research structure in the field of LTC is likely composed of multiple relatively independent clusters, which are internally cohesive but sparsely interconnected with 1 another.^[20]

Figure 4.

Author cooperation in the field of long-term care research.

4. Quantitative analysis of LTC research hotspots and frontier trends

4.1. Analysis of research hotspots based on keyword co-occurrence network

Word frequency analysis refers to the number of occurrences of characteristic words in the analyzed text data. Among them, keywords are highly condensed and summarize the main content of the paper. By proposing the keywords in the literature and conducting word frequency analysis, the hot spots and development trends in the research field can be explored and determined. In the keyword co-occurrence network mapping of LTC research, the top 5 word frequencies are LTC, dementia, nursing, health, and older adult (Table 3). They appeared 1666 times, 726 times, 674 times, 615 times, and 499 times, with centralities of 0.01, 0.02, 0.02, 0.01, and 0.02, respectively, indicating that LTC, dementia, and nursing are the most concerned topics in LTC research, and relevant studies are centered around them. Beyond the keyword “long-term care” itself, dementia and nursing home constitute the 2 most prominent key thematic priorities in this research field. The former (dementia) identifies cognitively impaired older adults as the primary care recipients, while the latter (nursing home) establishes institutional care settings as the predominant context for service provision. Furthermore,^[21] keywords such as health and older adult underscore the fundamental objective of this research: to enhance the health status and quality of life for the elderly population.

Table 3.

Top 20 high-frequency keywords in long-term care research.

Serial number	Keyword	Frequency	Centrality	Serial number	Keyword	Frequency	Centrality
1	Long-term care	1666	0.01	11	Quality of life	407	0.01
2	Dementia	726	0.02	12	Outcm	396	0.02
3	Nursing hm	674	0.02	13	Impact	382	0.01
4	Health	615	0.01	14	Nursing home	380	0.01
5	Older adult	551	0.01	15	Resident	368	0.01
6	Mortality	499	0.02	16	Management	353	0.01
7	People	498	0.01	17	Older people	352	0.01
8	Prevalence	487	0.02	18	Covid-19	307	0
9	Nursing home resident	436	0.03	19	Intervention	306	0.02
10	Risk	430	0.01	20	Long-term care facility	302	0.02

In order to further explore the hot issues in the field of LTC research, CiteSpace is used to cluster analyze the keywords of the WoS to obtain the literature since 2010. The frequency and centrality of keyword co-occurrence can reflect important indicators of research spots. In the generated keyword co-occurrence map, modularity Q (Q value) can be used to measure the significance of the cluster structure. It is generally considered that a Q value >0.3 indicates a significant cluster structure. The mean silhouette (S value) can be used to measure the similarity of the nodes within the clusters. It is generally believed that an S value >0.5 indicates a high degree of matching within the cluster and reasonable clustering.^[22]

By analyzing the co-occurrence clustering of the keywords, Figure 5 shows the keyword co-occurrence cluster map of the WoS core collection literature. The WoS map shows that the Q value = 0.3884 > 0.3 and S value = 0.744 > 0.5, which indicates that the clustering structure is significant and reasonable. Hot words with high co-occurrence frequency and medium centrality are: LTC, long-term mortality, LTC resident, LTC insurance, pragmatic trial, and social isolation. Early research primarily focused on the macro-level aspects of LTC, such as systemic frameworks, service delivery models, and quality metrics. With the deepening of scholarly inquiry, the field has progressively shifted its focus towards more specific and outcome-oriented issues, notably “long-term mortality” and “social isolation.” Concurrently, there is an active exploration of innovative solutions like “long-term care insurance” and rigorous “pragmatic trials” designed to address these challenges and ultimately improve outcomes for “long-term care residents.” This evolution signifies a strategic deepening of the field, moving from foundational questions of “what it is” to more applied investigations of “how to implement” and “how to improve.”

Figure 5.

Keyword co-occurrence clustering for long-term care research, 2010 to 2023.

Furthermore, the emergence of “social isolation” as a distinct and significant research cluster strongly indicates that the scope of LTC research has expanded beyond traditional concerns of physical illness and care management. It now explicitly prioritizes the psychological well-being and social connectedness of older adults as a core agenda. This shift aligns perfectly with the global emphasis on “healthy aging” and enhancing the overall “quality of life” within aging populations.

Transform the LTC research keyword co-occurrence cluster (Fig. 5) into a timeline view (Fig. 6). The timeline view, from top to bottom, indicates that the clusters are ordered from largest to smallest. For example, the cluster size of LTC is larger than long-term mortality. From left to right, the time sequence is represented; the leftmost data is 2010 and the rightmost data is 2024. The same horizontal line represents the same cluster, showing the evolution of a cluster over time. The keywords of LTC research from 2010 to 2024 are mainly divided into 7 clusters, which from largest to smallest are: LTC, long-term mortality, LTC residents, LTC insurance, utility trials, and social isolation. The largest clustering of LTC ranges from early studies of long-term palliative care and caregivers to residential aged care, person-centered care, surveys, moral distress, and so on. The second largest clustering of long-term mortality ranges from research on quality of life, risk, and guideline therapy to primary care, geriatrics, and atrial fibrillation by 2022 to 2024, public health, medication, etc. A close examination of the timeline view reveals several key characteristics of the field’s evolution. Firstly, and most notably, the research focus within each major cluster has demonstrably shifted over time, evolving from macro-level, foundational concepts towards more micro-level and specific issues. Secondly, the view illustrates a growing convergence of clinical medicine, public health, and the social sciences within LTC research. The discourse is no longer confined to the traditional domains of gerontology or nursing but actively incorporates more specialized clinical knowledge and a broader public health perspective. Finally, although the “Social isolation” cluster is not the largest in scale, it persists as a continuous thread throughout the timeline. It increasingly intersects with later concepts within the “long-term care” cluster, such as “person-centered care.” This indicates that the “psychosocial” experience in LTC has evolved from a peripheral concern into a core quality indicator and a legitimate target for intervention.^[23]

Figure 6.

Keywords in long-term care research (2010–2024) Timeline view.

4.2. Frontier analysis based on citation burst method

Citation burst can analyze the changes in keywords in research hotspots over time, that is, the active degree of keywords in a certain period of time, so as to understand the dynamics of research in the field. Through the emergence map in WoS, with LTC as the keyword, the keyword with the greatest emergent strength is COVID-19 (strength of 81.64), followed by loneliness (strength 10.89), unit (strength 9.07), index (strength 8.97), and so on. The biggest jumps in emergent time are setting (2012–2018), mini mental state (2012–2017), survival rate (2011–2015), and health insurance (2018–2020) (Fig. 7). From the perspective of burst intensity, COVID-19 demonstrates the highest value, indicating that the COVID-19 pandemic, as a public health emergency, rapidly became a focal point of research in the LTC field, particularly concerning the issues of infection, isolation measures, and mental health it brought about. The burst intensities of loneliness and unit rank second only to COVID-19, reflecting the sustained scholarly attention on “loneliness” and the “care unit” (or institutional setting) within LTC research. This also signifies the field’s ongoing concern for humanistic care and the organizational environment. Regarding burst duration, topics such as setting and mini mental state have demonstrated sustained prominence over many years, demonstrating that these research directions possess significant continuity and coherence. They likely correspond to fundamental issues in LTC, such as the care environment and psycho-cognitive status.^[24]

Figure 7.

Keyword emergence map in long-term care research, 2010 to 2024.

5. Analysis of hot areas of long-term care-related research

In order to avoid the 1-sidedness of dynamic network visualization analysis, it is generally necessary to combine it with the traditional combing method when using CiteSpace for scientific knowledge map analysis. In other words, on the basis of CiteSpace’s preliminary analysis of research hotspots, it is necessary to use the traditional combing method to specifically analyze the highly cited literature, in order to comprehensively and systematically reveal the developmental lineage and research frontiers. Scholars’ research about LTC can be traced back to the 1850s. This paper summarizes the literature from 5 aspects: theoretical research, model characteristics, research methodologies, impact mechanisms, and suggested strategies.

5.1. Theoretical research on LTC

Scholarly research on LTC has developed and enriched a number of theories, which effectively help practitioners and researchers to theorize and explain the phenomena and events that occur in LTC, and to explore causality, patterns, and regularities in the variables associated with LTC. For example, Rootedness Theory: Hunter et al used Rootedness Theory to find that the interaction between the elderly and caregivers is positive and bidirectional^[25]; Chiang et al used Rootedness Theory to study hospice care in LTC.^[26] Theory of change: De et al found that the change theory approach allows us to identify multiple intervention components for different stakeholders to achieve desired outcomes.^[27] Theory of Planned Behavior: Wang et al predicted the intentions and practices of caregivers in LTC facilities regarding physical restraints based on the theory of planned behavior.^[28]

5.2. The characteristics of LTC models

According to the World Health Organization, LTC for the elderly is divided into formal and informal care. Formal care is generally provided by nursing homes or specialized care facilities. Users of care services receive paid LTC services provided by service providers, individual caregivers, or family members at home, in particular, family members who must sign service agreements and receive remuneration for the provision of the service. Hlebec et al found that countries with lower levels of governance have less formal care provision, but the provision of formal services is increasing.^[29] This is closely related to the level of economic development of a country or region, as formal care is much more costly than informal care in terms of public aged care.^[30] Informal care refers to unpaid services provided by family members or friends in the family or community. Informal care is subdivided into institutional LTC and family LTC. Studies have found that the quality of life of the elderly in family care is lower than in institutional care,^[31,32] and family members mostly perceive caregiving as a burden.^[33,34] Therefore, the high cost of formal care needs to be coordinated by the government through public management, while the cost of informal care is low but the pressure on family members’ care is high.

5.3. The research methodologies in LTC

The research of LTC mainly consists of 3 research methods: qualitative research, quantitative research, and mixed research. In terms of qualitative research methods, Yu et al used the Delphi method to develop the LTC Literacy Assessment Scale for home caregivers in Taiwan.^[35] Jamwal used semi-structured interviews to study the engagement factors of health and LTC for the elderly in India.^[36] In terms of quantitative research methods, Guduk et al analyzed LTC preferences and their influencing factors by using Anderson behavioral model.^[37] Sato et al used logistic regression model to predict the demand for LTC for the elderly.^[38] Blüeher et al predicted the demand for LTC by multivariate statistical analysis of medical service assessment data.^[39] In terms of mixed methods, Adlbrecht et al conducted literature searches in PubMed, CINAHL, PsycINFO, Cochrane Library, and Web of Science databases, followed by comprehensive analysis of qualitative and quantitative studies.^[40] Liu et al analyzed 13 studies on the impact of intergenerational interactions on residents of LTC facilities in Asia.^[41] There are a growing number of ways to study LTC, ranging from macro to domain-specific studies.

5.4. The impact mechanism of LTC

The influence mechanism of LTC plays a key role in practical applications. Whether in policy development, resource allocation, theoretical research, or practical application, in-depth theoretical influence mechanisms are key to achieving goals, increasing efficiency, and improving outcomes. Jing et al propose that strong social relationships increase demand for formal care.^[42] Ismail et al propose that Turkey LTC and well-being model relies on families, especially women, to meet the increased demand. Expectations for LTC services are generally higher among the general population, especially among the elderly, women, those with family members requiring care, and those living in rural or regional areas.^[43] Courbage et al studied the main characteristics that influence adult children’s motivation and influence parents’ purchase of LTC insurance.^[44] Social relationships, care needs, and children’s motivations all have an impact on LTC, so researchers need to conduct multidimensional and multi-perspective studies to promote the development of LTC.

5.5. The suggested strategies for LTC

Scholars have conducted in-depth research in the field of LTC, offering valuable suggestions from different perspectives, which have enhanced the quality of decision-making and promoted the relevant practice of LTC. Salido et al recommend incorporating equitable financial compensation and social security benefits into innovative and sustainable strategies to support caregiving in LTC and welfare programs.^[45] Goncalves et al propose that with the increasing demand for LTC of the elderly, especially after the elderly are hospitalized and discharged, policies should be formulated to meet their LTC needs.^[46] Ghenta et al assess the LTC system in terms of sustainability, health, and quality of life for beneficiaries and their families, requiring good quality service coverage for LTC.^[47] Many scholars propose suggested strategies for LTC from macro, meso, and micro perspectives, providing useful references for policymakers.

6. Conclusions and prospects for the study

6.1. Conclusions

This paper uses bibliometric methods to visually analyze the WOS literature in the field of LTC research from 2010 to 2024. By drawing co-occurrence map, cluster map and emergent word map, this paper presents the keywords, authors and institution of LTC research literature and explores the hot topics and development trends of LTC research. The results are as follows:

First, in terms of contributing authors and research institutions, the author collaboration network exhibits an extremely low density. This finding closely aligns with the analysis of global LTC research by Fu et al,^[10] indicating that scholars in the LTC field predominantly tend to work independently or in small teams, lacking extensive and robust cross-team or cross-institutional academic exchange. The proportion of publications contributed by a core group of authors remains relatively low, suggesting that a dominant core author community has not yet been established, and collaborative interaction among authors is insufficient. LTC research is primarily concentrated within universities, with the main research institutions located in countries such as the United States, Canada, the United Kingdom, and Japan. While the institutional collaboration network density is higher than that of the author network, it remains at a significantly low level. This reflects a certain degree of collaborative relationships among publishing institutions, though the primary collaborative networks are limited in number and the strength of cross-institutional partnerships is generally weak. The data analyzed in this study, covering literature from 2010 to 2024, reveal that the trend of collaborative fragmentation has not significantly improved over time and may even have become more pronounced in certain dimensions. This provides a crucial benchmark for evaluating the effectiveness of academic community building within the field.

Second, from the analysis of keyword co-occurrence, the study found that all high-frequency keywords exhibit extremely low centrality (0.01–0.02). This reveals that while a substantial volume of research output in the LTC field revolves around a few core terms, the knowledge connections between these themes – and between them and broader research topics – remain weak. Consequently, the field is characterized by a fragmented state of “high productivity but low integration.” This observation aligns with the earlier finding of a low-density author collaboration network, indicating a dual deficiency in both scholarly collaboration and knowledge integration within the domain of LTC.

Third, from the analysis of keyword clustering and burst detection, the most salient finding is that COVID-19 emerged as the most prominent research hotspot during the 2010 to 2024 period with an overwhelming burst strength of 81.64. This fully corroborates and quantifies the conclusion drawn by Sun et al in their bibliometric study focused on pandemic-era LTC research: the COVID-19 pandemic fundamentally reshaped the research agenda in this field.^[11] However, the present study further specifies this impact by identifying high-burst keywords such as “unit” and “setting,” which reflect an intense, hyper-focused attention on the “care facility” as a critical micro-level entity. Furthermore, the emergence of “social isolation” as a distinct cluster in this study, alongside “loneliness” ranking as the second strongest burst keyword, extends beyond the frontiers (e.g., frailty, dementia care) identified in the 2019 review by Fu et al.^[10] It precisely captures a profound secondary effect brought about by the pandemic.

6.2. Prospects

First, at the macro-level: strengthening international cooperation and research, and complementing and learning from each other’s experiences and technologies. Through long-term practice and research, the valuable experiences and techniques of LTC have been summarized from all over the world in various countries and regions. Through collaborative research, advanced practices can be further promoted, combining the actual conditions of the country to enhance strengths and avoid weaknesses, and optimize the relevant policies and measures for LTC. Moreover, cooperation between different institutions should be strengthened. For example, universities, research institutions, medical institutions, communities, and governments should strengthen cooperation in scientific research; LTC research theories should guide the government and communities to optimize policies; and the practices of the government, communities, and medical institutions should be fed back to the research institutions to jointly promote the LTC system for the elderly.

Second, at the middle level: expanding the research of the elderly group and expanding the needs of the elderly. Analyzing the economic situation of different elderly groups and providing for the needs of the elderly groups according to their ability to pay. That is, elderly people in poverty and vagrancy should be taken into account, and the government and society should adopt appropriate policies to ensure the basic living care of the elderly and prevent the elderly from suffering due to financial difficulties. In addition, great attention should be paid to disabled and mentally disabled elderly groups, especially those with low or no income. Scholars need to conduct in-depth research in these aspects to ensure that low-income, disabled, and mentally disabled elderly people enjoy basic LTC policies.

Third, at the micro level: recognizing the increase of chronic diseases in the elderly and supporting the relevant medical needs of the elderly. With the development of society, the life expectancy of the elderly is extended, accompanied by an increase in the elderly population, and even some elderly suffer from several chronic diseases. Chronic diseases pose serious challenges to the life of the elderly, affecting their quality of elderly. In addition to medical aspects, it is also necessary for the government, society and the market to address the demand for services for the high prevalence of chronic diseases among the elderly.

Acknowledgments

Data from a publicly available Web of Science were used in this study, and the authors would like to thank all those who contributed and participated in the data collection.

Author contributions

Conceptualization: Xianlin Bi, Qingquan Pang.

Data curation: Qingquan Pang.

Investigation: Shihua Xu.

Methodology: Shihua Xu, Yue Li.

Project administration: Yue Li.

Resources: Yun Zhao, Yue Li.

Software: Yun Zhao.

Supervision: Yongping Nong, Haidan Qin.

Validation: Xianlin Bi, Yongping Nong.

Visualization: Xianlin Bi, Haidan Qin.

Writing – review & editing: Xianlin Bi.