Bibliometric Analysis of Hospital Design: Knowledge Mapping Evolution and Research Trends

Abstract

Hospital design plays a pivotal role in improving patient outcomes, enhancing clinical efficiency, and strengthening infection control. Since the outbreak of COVID-19, research in this field has expanded significantly, showing a marked trend toward interdisciplinary integration. In this study, bibliometric analysis was conducted using CiteSpace (version 6.2.R3) as the primary tool, with Excel and Tableau (version 2024.3) as supplementary software. A total of 877 documents on hospital design published between 1932 and 2025 were retrieved from the Web of Science Core Collection and analyzed from multiple perspectives. The analysis examined publication trends, collaborative networks, co-citation structures, disciplinary evolution, and keyword dynamics. The results indicate that the field has entered a phase of rapid development since 2019. Global collaboration networks are becoming increasingly multipolar; yet, institutional and author-level connections remain decentralized, with relatively low overall density. Evidence-based design (EBD) continues to serve as the theoretical foundation of the field, while emerging themes such as healing environments, biophilic design, and patient-centered spatial strategies have become major research hotspots. Increasingly, the field reflects deeper integration across disciplines, including architecture, medicine, nursing, and environmental science. This study provides a clearer picture of the developmental trajectory, knowledge base, and future directions of hospital design research, offering systematic insights and theoretical guidance for both scholars and practitioners.

1. Introduction

In recent years, hospital design has emerged as a critical interdisciplinary topic at the intersection of architecture, medicine, public health, and environmental psychology [1,2,3]. Hospitals are not merely spatial carriers for diagnostic and therapeutic functions but also complex systems that impact patient recovery, healthcare worker well-being, and public health safety [4,5,6]. The World Health Organization (WHO) [7] and various national governments have published design standards and evaluation systems that address functional layout, infection control, perceptual experience, and sustainability. These efforts highlight the central role of hospitals in addressing global health challenges. Global health crises, including the Ebola outbreak in 2014 and COVID-19, have highlighted the close connection between public health and architectural design [8,9]. Hospital design is now a key component of public health resilience systems—adaptive frameworks that enable healthcare facilities to withstand and recover from crises [10]. This role is especially critical in responding to public health emergencies.

Hospital design is undergoing a profound transformation with the rise of human-centered design and the application of data analytics. It has shifted from empirical induction to evidence-based methods and from efficiency-driven to human-centered approaches [11]. The representative “EBD” theory has fostered logical connections between design interventions and health outcomes [12,13,14]. Additionally, nursing, environmental science, and architectural engineering fields are also incorporating spatial factors into health-related research, fostering the formation of highly interdisciplinary research networks [15,16,17].

Despite the growing number of studies, several issues remain: uneven regional development, a concentration on specific hospital types, limited research on specialized facilities, and insufficient interdisciplinary integration. These gaps hinder the dissemination of findings and the advancement of theory [18]. There is still no systematic assessment of the long-term evolution of the knowledge system in medical building design. In particular, horizontal reviews of global cooperation networks, thematic evolution, and disciplinary integration models are missing [19]. Another important gap is the lack of systematic reviews on how long-term epidemic events shape the knowledge structure. Addressing this requires multidimensional methods to reveal evolutionary pathways, which represent the dynamic trajectories of knowledge development and thematic change across different periods [20].

Building on the above overview of hospital design, this study systematically examines the evolution of research across the full period from 1932 to 2025 using bibliometric methods. Bibliometric analysis is a quantitative approach that applies mathematical and statistical techniques to scientific publications in order to reveal knowledge structures, collaboration patterns, and research trends. Knowledge mapping, in this context, refers to the visualization of relationships among publications, authors, and concepts, which helps to clarify the intellectual structure of the field. Research trends denote the evolving directions of scholarly output over time [21,22]. This study employed CiteSpace (version 6.2.R3) as the primary analytical tool, supplemented by Excel and Tableau (version 2024.3) for auxiliary analyses. The investigation covers multiple dimensions, including publication trends, national and institutional collaboration networks, author and journal co-citation structures, interdisciplinary collaboration patterns, keyword clustering, and emerging research fronts. For hospital design, bibliometric methods are especially valuable because they enable researchers to systematically trace long-term trends, identify influential works, and visualize the dynamic evolution of interdisciplinary knowledge—capabilities that are difficult to achieve through traditional narrative reviews. Previous bibliometric studies in related domains have further demonstrated the effectiveness of these methods in mapping research landscapes. For instance, bibliometric analyses have been applied in architecture and urban planning [23,24], healthcare architecture [19], and broader domains of environmental design and planning [25]. However, a comprehensive bibliometric study focusing specifically on hospital design has not yet been conducted. This study addresses the dynamic evolution of research hotspots, collaboration structures, and knowledge systems in hospital design. Its objectives are to construct a systematic knowledge map, reveal the evolutionary trajectory of the field, and provide interdisciplinary, human-centered, and resilient support for future hospital design. Theoretically, it illustrates how the knowledge system centered on evidence-based design (EBD) has developed over time and generated new research branches, thereby filling the gap in systematic reviews of the century-long evolution of hospital design and its knowledge structure. Methodologically, it validates a multidimensional research paradigm based on bibliometric mapping and establishes a framework that integrates knowledge graph analysis with visualization techniques. Practically, it contributes to evidence-based, human-centered, and adaptable decision-making for hospital and healthcare facility design, supporting the creation of more resilient and patient-centered medical spaces in the future.

The following part of this paper (Section 2) reviews the relevant literature in the field of medical building design. Section 3 introduces the research methods and data processing strategies. Section 4 introduces the research results, including research trends, collaboration networks, the core literature, and research hotspots. Section 5 discusses the research results. Section 6 summarizes the contributions of this study and future research directions.

2. Literature Review

As the focus on hospital design continues to grow, related policies, regulations, and design guidelines are also being introduced and updated. National and international guidelines have played a crucial role in facilitating the dialogue between design practice and research. In the WHO’s continuously updated guidance documents on hospitals, the emphasis on the flexibility of medical spaces, psychological support, and infectious disease prevention and control functions has gradually become an increasingly important component [26,27,28]. The U.S. Facility Guidelines Institute (FGI) issued the “Hospital Facilities Design Code”, and the U.S. Department of Health and Human Services (HHS) revised relevant chapters in response to the COVID-19 pandemic in 2025 [29,30]. The UK’s National Health Service (NHS) has provided comprehensive standards covering sustainability, inclusivity, and safety [31]. Additionally, countries such as Australia [32] and China [33] have successively introduced hospital design policy documents tailored to regional contexts, emphasizing green building standards, seismic resistance, and information infrastructure.

The theory of hospital design has also undergone multiple stages of development alongside the strengthening of public health systems. Starting from the functionalist design philosophy of the early 20th century, hospitals were viewed as technical spaces characterized by efficient circulation, standardized processes, and zoned management. After the 1970s, with the rise of patient rights awareness and humanistic care concepts, design began to shift its focus toward the experience of patients and healthcare workers, doctor–patient relationships, and psychological rehabilitation processes [34,35,36,37,38]. In 1984, Roger Ulrich’s classic study, “View through a Window May Influence Recovery from Surgery”, was widely regarded as the starting point of EBD [39], a theory that emphasizes the empirical impact of healthcare spaces on clinical outcomes, gradually expanding to encompass multiple dimensions, such as environmental interventions, perceptual experiences, and care quality [40,41]. This developmental trajectory also includes a transition from classic “window view” experiments to large-scale pilot studies of multivariate environmental interventions [42,43,44,45]. Subsequent studies have delved into topics such as the role of ventilation systems in infection control [46], recovery space design for a safer rehabilitation environment [47,48,49], the impact of healthcare spaces on patient satisfaction and psychological recovery [50,51,52], the optimization of intensive care unit spaces [53,54,55], the mechanisms shaping rehabilitation environments [56,57], and the sustainability of green hospital construction [58,59,60]. More recently, concepts like “healing architecture [61,62,63]” and “salutogenic design [64,65]” have been gaining traction in healthcare settings, emphasizing promotion rather than disease prevention and control [66]. Additionally, some studies have proposed the concept of “modular resilience” to support rapid responses to sudden public health emergencies [67,68,69,70]. During the COVID-19 pandemic, modular isolation wards, mobile negative-pressure rooms, and digital screening kiosks were rapidly deployed [71,72,73,74,75]. The pandemic has significantly heightened international focus on the adaptability and resilience of healthcare spaces. Hospitals faced the urgent need for rapid transformation, requiring them to address multiple challenges within a short timeframe, including increasing isolation spaces, renovating ventilation systems, expanding emergency areas, and supporting the mental health of patients and healthcare workers [76,77,78,79]. The crisis prompted a reevaluation of the limitations of traditional function-oriented hospital design. By enhancing the resilience and improving crisis response mechanisms, research has increasingly focused on dimensions such as user experience, environmental psychological interventions, and performance evaluation mechanisms [80,81]. The pandemic has not only altered the functional requirements of hospitals but also accelerated a fundamental shift in hospital design from “reactive response” to “data-driven preventive construction”, driving the field to transition from an “efficiency-oriented” approach to a “human-centered” one [82].

While theoretical developments and emerging trends evolve, research methods and technical approaches in hospitals have shown a trend toward interdisciplinary integration. While traditional methods primarily relied on questionnaires, interviews, and case studies, recent research has employed behavioral mapping, spatial syntax, CFD simulation, VR simulation, and post-occupancy evaluation (POE) [83,84,85,86]. Technologies like BIM, digital twins, and machine learning have also been applied to spatial performance analysis and layout optimization. With the assistance of Bayesian models and machine learning algorithms, architectural researchers can now simulate the impact of patient flow patterns and airflow diffusion on medical efficiency and infection risks [87,88,89]. Notably, bibliometric analysis, which empowers medical space research by supporting knowledge structure organization, performance evaluation, and design optimization, has emerged as a systematic research tool in this field [90]. Bibliometric analysis has become an established method for quantitatively evaluating the knowledge base and research dynamics of a scientific field. It employs indicators such as co-citation, co-authorship, and keyword co-occurrence to uncover the intellectual structure, collaborative patterns, and thematic evolution of research domains. A variety of software tools have been developed to facilitate bibliometric analysis, including VOSviewer, CiteSpace, and Bibliometrix, each providing unique strengths for mapping networks and visualizing research trends [91]. The bibliometric analysis offers an appropriate and necessary approach for systematically examining the long-term evolution of hospital design research.

In recent years, bibliometric methods have been increasingly applied to healthcare and hospital-related research, offering insights into thematic development and methodological gaps. For example, Yang et al. (2021) [92] analyzed the global literature on Magnet Hospitals and identified research hotspots in nursing practice, patient safety, and organizational outcomes. Philip et al. (2023) [93] conducted a bibliometric review of simulation modeling in hospital outpatient departments, highlighting methodological advances and the growing application of system-based approaches in healthcare operations. Nepomuceno et al. (2022) [94] performed a bibliometric analysis of frontier efficiency studies in hospitals, showing how efficiency metrics have been used to assess healthcare performance. Singh and Ravi (2023) [95] mapped the application of Lean Six-Sigma in hospitals over the past decade, identifying productivity improvement and quality assurance as central themes. Wahyuni et al. (2024) [96] examined bibliometric trends in technology use and innovation in healthcare services, underscoring the increasing role of digitization and innovation in hospital contexts. Collectively, these studies demonstrate the potential of bibliometric analysis to uncover research structures, thematic focuses, and methodological patterns in hospital-related fields. However, despite the importance of hospital design as a key dimension of healthcare infrastructure, to date, no bibliometric study has comprehensively addressed this domain. This gap highlights the originality and necessity of the present research.

Despite the abundance of existing research findings, several issues warrant attention. First, regional disparities persist, with most studies dominated by developed Western countries, while localized strategies and culturally responsive designs in underrepresented regions are lacking [97]. Second, the types of hospital buildings studied remain limited, with insufficient focus on psychiatric hospitals, maternity centers, or post-disaster temporary facilities [98,99]. Third, interdisciplinary integration is still underdeveloped, as gaps persist between architectural design and clinical medicine or environmental psychology [100]. In this context, future research in hospital design should prioritize comparative empirical studies across multiple countries, the strengthening of interdisciplinary integration methods, the development of mixed assessment models for real-world scenarios, and the construction of resilient hospital design frameworks to address public health emergencies.

3. Materials and Methods

Knowledge mapping, as a cutting-edge analysis method in bibliometrics, can visualize and express the quantitative analysis results of subject knowledge [20]. CiteSpace is a knowledge mapping software based on the Java platform, developed by Chaomei Chen [101]. It has been widely applied in more than 60 research fields due to its capacity to analyze co-occurrence patterns and collaborative networks within database documents and to visually represent evolutionary trends, knowledge associations, and emerging research frontiers [102]. The bibliometric methodology was used to systematically analyze the research development in the hospital design field using CiteSpace (version 6.2.R3) as the primary tool in this study.

3.1. Data Collection

The data were obtained from the WoSCC, which provides bibliographic records from 1900 to the present. The search strategy was designed through an iterative process, in which multiple synonyms and related terms for “hospital” were tested. After trial retrievals, the final query was determined as: TS = (“hospital design” OR “healthcare facility design” OR “medical center design”) applied to the Topic field (title, abstract, author keywords, and Keywords Plus) in the Web of Science Core Collection. This formulation ensured the maximum coverage of relevant studies while avoiding redundancy among overlapping terms. To ensure comprehensiveness, we included the available WoSCC sub-databases shown in the Clarivate platform interface, including Science Citation Index Expanded (SCI-EXPANDED, coverage from 1900–present), Social Sciences Citation Index (SSCI, 1956–present), Conference Proceedings Citation Index-Science (CPCI-S, 1990–present), Conference Proceedings Citation Index-Social Science & Humanities (CPCI-SSH, 1990–present), Current Chemical Reactions (CCR-EXPANDED, 1985–present), and Index Chemicus (IC, 1993–present). As emphasized by Liu (2019) [103], explicitly disclosing the sub-databases and their coverage years is essential for transparency and reproducibility in bibliometric research. The search was conducted on 20 January 2025, covering the full WoSCC timespan (1900–2025), with retrieved records spanning 1932 to 2025. In order to achieve the most comprehensive coverage of the field, no additional restrictions were imposed on the document type. This inclusive strategy allowed the retrieval of the broadest possible set of relevant studies. All 877 relevant original articles remained and were exported as plain text in the format of ‘Full Record and Cited References’ for CiteSpace (version 6.2.R3) graph construction and statistical visualization analysis. Duplicate records were removed using the built-in deduplication function in CiteSpace (version 6.2.R3) during data import, which automatically detects and excludes identical entries. To reduce noise in collaboration and co-citation networks, author and institution names were standardized by combining CiteSpace’s built-in name-merging function with manual correction.

3.2. Bibliometric Tools and Settings

This study used CiteSpace (version 6.2.R3), Excel, and Tableau (version 2024.3) to improve the multidimensionality and precision of the analysis. CiteSpace (version 6.2.R3) was used to construct and cluster the literature network, and the CiteSpace parameter settings are as follows [104]:

• Setting time period as 1932–2025 (the full time period), with 1 year per slice, where each slice selected 50 high-frequency nodes (Top N = 50);
• Adjusting the k-value flexibly to the requirements of different analysis types to optimize the clarity and structural presentation of the graph based on the g-index algorithm;
• Using the ‘Pathfinder’ and ‘Pruning the merged network’ settings for network structure pruning;
• Using the Log-Likelihood Ratio (LLR) clustering algorithm for cluster analysis to extract noun terms from the titles of these documents to name the research contents corresponding to the clusters [105].

3.3. Visualization and Network Construction

Quantitative indicators from CiteSpace (e.g., frequency, Betweenness Centrality, and burst detection) were combined with qualitative domain-informed reading to interpret networks and clusters. Two standard metrics, i.e., Modularity Q and Silhouette S, were employed to validate the structural significance and robustness of the generated clusters and to guide the interpretation of the knowledge mapping outputs. The Modularity Q index (ranging from 0 to 1) measures the degree to which a network can be divided into distinct communities; values above 0.3 generally indicate significant modular structures. The Silhouette S value (ranging from −1 to 1) evaluates the internal consistency of clusters; values above 0.5 suggest reasonable clustering, while values above 0.7 indicate highly reliable results [20,21]. LLR-derived labels guided the interpretation of thematic content and its evolutionary pathways. Where relevant, we contextualized patterns across historical phases (e.g., the acceleration of output after the COVID-19 pandemic) to link mapped structures with developments in hospital design scholarship. A variety of visualizations were generated by CiteSpace (version 6.2.R3), including network maps, time zone views, timeline views, clustering networks, mountain views, etc. We constructed the following:

• Collaboration networks at country/region, institution, and author levels to reveal cooperation structures;
• Co-citation networks of authors and journals to identify influential intellectual bases;
• Reference co-citation and disciplinary co-occurrence maps to outline knowledge structure and disciplinary distribution;
• Keyword co-occurrence maps to capture research themes and hotspots.

Excel and Tableau (version 2024.3) assisted in the statistical analysis of the trends in the number of publications and the distribution of countries and regions, complementing the CiteSpace (version 6.2.R3) visual networks.

3.4. Interpretation Strategy

First, we used Excel and Tableau (version 2024.3) to plot annual publication counts and to map the global distribution of outputs by country/region, thereby identifying temporal and regional trends in hospital design research. Next, we constructed knowledge maps in CiteSpace (version 6.2.R3). We examined the structure of the research community by generating co-occurrence networks of countries, institutions, and authors. We also conducted a focused review of highly cited papers to capture the seminal contributions that shaped the field. We set a threshold of ≥100 citations to clearly separate the most influential works from the long tail of citation counts; 13 papers met this criterion. These documents were analyzed qualitatively, paying attention to their research themes, methodological contributions, and downstream influence on hospital design studies. We then produced author and journal co-citation networks to identify influential scholars and venues and analyzed knowledge structures and disciplinary distributions using reference co-citation and co-occurrence maps. Finally, we examined keyword co-occurrence and bursts to characterize research themes, hotspots, and emerging fronts. A step-by-step overview of the research design is provided in Figure 1.

It should be noted that, as with most bibliometric studies, this methodology has certain limitations. The use of the WoSCC as the sole data source may introduce language bias (given its emphasis on English-language journals), which may limit the visibility of regional or non-English scholarship and exclude the grey literature, such as reports and policy documents [106,107]. In addition, differences or overlaps between the WoSCC and other databases (e.g., Scopus and PubMed) may also influence coverage. These inherent constraints are acknowledged here for transparency and are further elaborated in the Limitationsof Section 5 (Discussion).

4. Results

4.1. Research Trends and Overview of Research Groups

4.1.1. Analysis of the Number of Publications

The number of publications is an important indicator of research trends. Figure 2 illustrates the annual distribution of publications on hospital design, based on WoSCC records as of 20 January 2025.

Before 1998, the research output in this field was sparse, and the overall number of publications remained low. This indicates that hospital design had not yet become a widely recognized academic topic. From 1998 to 2019, the number of publications began to rise steadily, reflecting the gradual establishment of hospital design as a distinct area of scholarly interest. Between 1 January 2020 and 20 January 2025, the number of related publications rose to 417 (47.56% of the total), which is more than double the increase observed in 1998–2019. While related healthcare and design domains also experienced growth after COVID-19, bibliometric studies [95,96] indicate that the surge in hospital design is particularly pronounced.

It should also be noted that the expansion of WoSCC coverage contributed to this growth. The inclusion of new datasets may artificially inflate counts in some periods, and sudden surges may partly reflect changes in database policies rather than purely academic momentum [108,109]. Thus, the observed growth is likely the result of both database expansion and genuine research activity. The fitted trend line (R² = 0.6348) indicates an overall upward trajectory of scholarly interest [110]. In particular, after 2020, the surge in publications demonstrates that hospital design has become a heated research focus with increasing international attention.

In summary, publication activity in hospital design research can be divided into three stages: (1) a dormant phase before 1998, (2) steady growth from 1998 to 2019, and (3) rapid expansion after 2020. This trajectory confirms the rising academic significance of the field and its heightened relevance in the context of global health challenges.

4.1.2. Analysis of the Cooperation Network of Countries

The country cooperation network, shown in Figure 3, characterizes the collaborative relationships and network structure between countries in the field of hospital design. It should also be noted that many older records in the WoSCC lack complete author address information, as documented in bibliometric methodology studies [111]. This absence may partly explain the sparse representation of certain countries in earlier decades and should be considered when interpreting long-term collaboration patterns.

Table 1. Top 5 countries contributing to hospital design. (Note: Records originally indexed as “England” in the WoSCC have been standardized as “United Kingdom (UK)”).

No.	Country	Count	Centrality
1	USA	260	0.17
2	AUSTRALIA	86	0.17
3	United Kingdom (UK)	82	0.2
4	PEOPLE’S R CHINA	43	0.13
5	CANADA	40	0.08

presents the top five countries by publication volume. Combined with the collaboration mapping, these data reveal the geographic distribution of research output and the cross-regional cooperation patterns that characterize the field of hospital design. The relative influence of countries was assessed through Betweenness Centrality, where nodes with higher values indicate greater bridging roles in international collaboration [112]. In CiteSpace visualizations, countries with high centrality are highlighted by purple circles, reflecting their stronger research impact. And the size of the circle corresponds to the magnitude of the Count value.

In summary, the country cooperation network shows that global participation in hospital design research is uneven. A few countries dominate the field through both publication output and centrality, while incomplete metadata in older records have likely contributed to the apparent underrepresentation of others.

The global distribution of publications on hospital design shows clear regional disparities (Figure 4). Overall, the United States is the most dominant research country in the field, topping the list with 260 publications. Its Betweenness Centrality is 0.17, with a dense and extensive connectivity line centered on the USA node radiating to countries such as the United Kingdom (UK) and Australia. It shows its dominance in global hospital design research and its pivotal role in the global collaborative network. Australia and the UK ranked second and third, with 86 and 82 publications, respectively, showing strong research strength and continuous research investment, but their network structure roles are slightly different. The UK functions as a knowledge transit intermediary connecting Commonwealth countries, European countries, and the United States. Australia, on the other hand, is prominent in connectivity within the Asia-Pacific region.

Other active contributors include China (43 articles) and Canada (40 articles), which also have significant research contributions. Notably, Japan shows an exceptionally high Betweenness Centrality (0.63), serving as a strategic bridge between the US and Asian countries. Similarly, Turkey (0.43) links Europe with West Asia, while Italy (0.38) strengthens collaboration paths with China and Germany. These countries contribute less in volume but compensate through high centrality, frequently appearing as “bridges” in the cooperation network.

Several European countries, such as the Netherlands, Denmark, Italy, and Sweden, along with India, each produced more than 20 publications, reflecting sustained participation in international hospital design research.

In contrast, South America and Africa remain underrepresented. Only a few countries, such as Brazil (11), Egypt (17), Nigeria (7), and South Africa (7), show limited activity, while most countries in these regions remain scarcely studied.

In summary, hospital design research is geographically concentrated. The US, the UK, and Australia dominate in terms of output, with China and Canada also contributing actively, while countries such as Japan, Turkey, and Italy play bridging roles due to high centrality. Large parts of South America and Africa remain marginalized, highlighting global disparities in research attention.

4.1.3. Analysis of the Cooperation Network of Institutions

The cooperation network of institutions, as shown in Figure 5, presents the characteristics of the cooperative relationship and network structure among the major research institutions in hospital design. This helps to analyze the influence of the institutions in this field and encourages them to strengthen the cooperation to promote the development of the research field [113].

Based on node size and Betweenness Centrality, the University of London, Aarhus University, and the University of Toronto emerge as the most influential institutions. These universities contribute substantially to publication output and play bridging roles in connecting different research communities. Notably, Aarhus University has the highest centrality (0.18), confirming its role as a pivotal connector across clusters.

The network also reveals several geographically defined clusters. Collaboration-intensive clusters divided by geography are evident in Australia (consisting of the University of Melbourne, Monash University, Deakin University, etc.), North America (consisting of Harvard University, the University of Toronto, the US Department of Veterans Affairs, etc.), and Europe (consisting of KU Leuven, Erasmus MC, Erasmus University Rotterdam, etc.), with the clusters showing closer cooperation within them.

In summary, institutional collaboration in hospital design research is characterized by a few highly central universities and regionally concentrated clusters. While these institutions drive productivity and connectivity, the overall network still shows limited integration across continents, suggesting a scope for broader international collaboration.

4.1.4. Analysis of the Cooperation Network of Authors

The core authors in hospital design were identified based on the Price formula (N = 0.749(Nmax)1/2), where Nmax is the number of articles published by the author with the most publications in the field [114]. The author cooperation network (Figure 6) reflects collaborative structures between 1932 and 2025.

A total of 713 author nodes and 767 collaboration lines were identified, with a network density of 0.003. This low density indicates a highly decentralized collaboration pattern, where small author groups dominate rather than having a unified research community.

Several clusters display stable collaboration patterns, such as the group of Blanc B., Cravello L., and Roger V., as well as the long-term partnership between Curtis, Sarah E. and Close, Helen J. While some authors have relatively high publication counts, their roles as network connectors are limited.

Indeed, the majority of authors have a Betweenness Centrality of 0, suggesting weak bridging capacity between clusters. A few authors, such as Noskin G.A. and Peterson L.R., are active within specific groups but lack broader cross-group collaborations.

The temporal evolution of node colors reveals another important feature: weak links between earlier researchers and more recent contributors. This indicates a clear generational disconnect in author collaborations, with few enduring connections across different research periods.

In summary, the author cooperation network is fragmented, characterized by multiple small groups, limited cross-team collaboration, and a lack of strong bridging authors. The weak intergenerational links further highlight the need for broader and more sustained collaboration in this field.

4.2. Identification of Core Contributions and Influence

4.2.1. Analysis of Reference Citations

The number of reference citations, as an important indicator of the academic impact of the literature, can reflect the breadth of dissemination and recognition. The representative problem domains, research paradigms, and theoretical sources in the field can be seen by combing the highly cited references [115]. Based on the WoSCC dataset, citation counts were extracted and ranked. In total, 13 references with more than 100 citations were identified and analyzed; they are summarized in

Table 2. References with more than 100 citation counts.

No.	Author	Title	Citation Count
1	Ulrich et al. (2008) [39]	A systematic overview of the positive impact of the built environment on patient recovery, healthcare outcomes, and employee performance; a seminal text in EBD.	828
2	Huisman et al. (2012) [63]	Assessed the impact of physical environmental factors such as light, noise, and air quality on the physical and mental health and quality of care of hospitalized patients, emphasizing the critical role of environmental variables in healthcare design.	293
3	Priestley et al. (2004) [116]	Empirical support for clinical interventions and spatial coordination through a field trial examining the effectiveness of ICU outreach services in improving the efficiency of critical care management.	250
4	Shaughnessy et al. (2011) [117]	An empirical analysis of the impact of ward allocation patterns on the transmission of Clostridium difficile infection, emphasizing the effectiveness of single-occupancy wards in infection control.	238
5	Lankford et al. (2003) [118]	Investigated the effects of role modeling and hospital spatial layout on healthcare workers’ hand hygiene behavior to reveal the interactive mechanisms between behavioral interventions and environmental factors.	200
6	Gesler et al. (2004) [119]	Evaluated the impact of UK hospital architectural reforms on patient experience and outcomes and proposed a patient-centered design orientation.	172
7	Ulrich et al. (2010) [120]	A systematic conceptual framework for EBD was proposed to establish a theoretical foundation and practical assessment logic for healthcare building research.	171
8	Warshaw et al. (1982) [121]	Early exploration of spatial adaptations for dysfunction in hospitalized older adults provides important clues to the design of healthcare spaces for special populations.	152
9	McGain et al. (2014) [122]	Summarized environmental sustainability topics in hospital buildings and proposed a strategic framework and research agenda for green healthcare design.	134
10	Curtis et al. (2007) [123]	Assessed the healing landscape and spatial perceptual effects of a new psychiatric ward based on a psycho-geographic perspective, emphasizing the importance of the subjective experience of the user.	129
11	Lip et al. (1994) [124]	Explored the efficiency of room configurations in intervening in the treatment of atrial fibrillation in a general hospital, reflecting the role of ward layout in supporting specific clinical processes.	124
12	Chaudhury et al. (2005) [125]	Summarized the advantages and limitations of single- versus multi-occupancy wards in terms of privacy, infection control, and patient satisfaction.	119
13	Dalke et al. (2006) [126]	Investigated the psychological and physiological roles of color and lighting in the hospital space and proposed a design approach to optimize the hospital experience with sensory modulation.	116

.

These highly cited works were published between 1982 and 2014, with the majority concentrated after 2000. Most appeared in high-impact journals related to the built environment, healthcare, and infection control. The distribution of these references underscores the cross-disciplinary nature of hospital design research. It shows that the field, situated at the intersection of architecture and medicine, has gradually attracted significant academic attention since the early 21st century.

The most highly cited paper was authored by Ulrich et al., with 828 citations, which is widely recognized as a seminal work in evidence-based design (EBD). Subsequent co-citation analyses and the identification of the emergent literature further confirm its central significance to the field.

The research themes of the highly cited literature show a clear concentration. Many studies emphasize the positive impact of spatial configuration on efficiency, safety, and user experience. Core topics include healing environments, infection control, and critical care system design, reflecting the dual orientation of hospital design toward both enhancing patient recovery and preventing disease transmission.

Several earlier publications continue to maintain high citation counts. This persistence suggests that the theories and methods introduced have long-term continuity and adaptability, serving as foundational pillars of hospital design research.

Most of the highly cited works were published between 2004 and 2012, marking a critical period of transition from empiricism to evidence-based healthcare architecture. During this time, research addressed not only classical themes, such as healing environments and design optimization, but also emerging concerns related to epidemic prevention and crisis response. This illustrates the field’s responsiveness to evolving social and public health challenges.

In summary, the analysis of highly cited references highlights both the enduring influence of classical theories and the dynamic emergence of new themes. Together, these works have shaped hospital design into a field that balances patient-centered healing with resilience against public health crises.

4.2.2. Analysis of the Co-Citation Network of Authors

Figure 7 shows the structure of key academic figures and knowledge networks in the hospital design field. The overall structure shows a loose polycentric aggregation structure, with close co-citation relationships among multiple authors, especially in the clusters of “EBD”, “healing environment”, etc. The color gradient of the nodes reflects the time dimension of the literature citations, and the transition from green to yellow indicates that most of the core authors’ studies were concentrated after 2000, suggesting that the research hotspots in this field have been gradually formed and deepened in the past two decades.

The information in

Table 3. Top 10 co-citation authors. (Note: “Year” refers to the earliest publication year of the cited author’s work recorded in the WoSCC).

No.	Author	Count	Centrality	Year
1	ULRICH RS	193	0.2	1991
2	WORLD HEALTH ORGANIZATION	42	0.03	2019
3	PATI D	40	0.01	2011
4	JOSEPH A	38	0.15	2008
5	CHAUDHURY H	32	0.23	2007
6	SHEPLEY MM	31	0.09	2010
7	HAMILTON DK	29	0.19	2008
8	HUISMAN ERCM	21	0.08	2017
9	MABEN J	21	0.06	2018
10	TAYLOR E	20	0.01	2020

shows that ULRICH RS is the author with the highest frequency of co-citation in the field (193 citations, Betweenness Centrality = 0.20), indicating he has a bridging role between several research groups and that his research is of great importance. JOSEPH A and CHAUDHURY H also have high visualization weight and centrality in the network, indicating their research results have played an important role in supporting the development of the field. The WORLD HEALTH ORGANIZATION (WHO) is also frequently co-cited, especially in global public health cases in recent years, reflecting the fact that the design and management guidelines issued by it are widely cited. In addition, authors such as HAMILTON DK, PATI D, MABEN J, and SHEPLEY MM are at the heart of the network, forming a major knowledge base in hospital design.

In summary, the author co-citation network highlights the central role of a few key scholars and organizations in shaping the field. Their work has not only established evidence-based foundations but also guided the thematic evolution of hospital design over the past two decades.

4.2.3. Analysis of the Co-Citation Network of Journals

CiteSpace (version 6.2.R3) was used to identify the distribution of key journals in hospital design research based on co-citation analysis. Journal distribution is a crucial indicator of the development trends of a field, as it reflects the main publication outlets and their influence. The co-citation network was assessed through node frequency, link strength, and density to generate a ranking of influential journals [127]. Figure 8 presents the journal co-citation network, with detailed co-citation frequency and Betweenness Centrality values listed in Table 4. The analysis highlights several journals with high co-citation frequency and centrality:

• The Health Environments Research & Design Journal (HERD) is the journal with the highest co-citation frequency in this field, reflecting its importance as a specialized journal in the hospital design field.
• The Lancet and New England Journal of Medicine (NEJM) are authoritative journals in the medical field, which also play an important role in supporting the literature on the research of medical buildings.
• The Journal of the American Medical Association (JAMA) shows a high centrality, indicating that it has a sound knowledge intermediary role between different research groups.
• The Journal of Advanced Nursing and Journal of Clinical Nursing show the contribution of the nursing field to healthcare space.
• Environment and Behavior is the journal with the highest co-citation centrality, indicating that it has a significant bridging role in connecting environmental psychology and medical building research.
• PLOS ONE, the British Medical Journal (BMJ), the Journal of Environmental Psychology, and other journals also have a high co-citation frequency, reflecting the obvious interdisciplinary characteristics of this field, involving multiple directions, such as medicine, psychology, environmental behavioral science, and architecture.

Figure 8. Journal co-citation network.

Figure 8. Journal co-citation network.

Table 4. Top 10 co-cited journals. (Note: “Year” refers to the earliest year in which the cited journal’s publications appeared in the WoSCC within the dataset).

No.	Cited Journals	Count	Centrality	Year
1	HERD-HEALTH ENV RES	226	0.02	2010
2	LANCET	140	0.08	1987
3	JAMA-J AM MED ASSOC	121	0.11	1995
4	J ADV NURS	112	0.1	2004
5	NEW ENGL J MED	102	0.15	1999
6	ENVIRON BEHAV	94	0.21	2006
7	PLOS ONE	91	0.02	2013
8	BMJ-BRIT MED J	85	0.1	1989
9	J ENVIRON PSYCHOL	76	0.06	2011
10	J CLIN NURS	75	0.07	2013

Table 4. Top 10 co-cited journals. (Note: “Year” refers to the earliest year in which the cited journal’s publications appeared in the WoSCC within the dataset).

No.	Cited Journals	Count	Centrality	Year
1	HERD-HEALTH ENV RES	226	0.02	2010
2	LANCET	140	0.08	1987
3	JAMA-J AM MED ASSOC	121	0.11	1995
4	J ADV NURS	112	0.1	2004
5	NEW ENGL J MED	102	0.15	1999
6	ENVIRON BEHAV	94	0.21	2006
7	PLOS ONE	91	0.02	2013
8	BMJ-BRIT MED J	85	0.1	1989
9	J ENVIRON PSYCHOL	76	0.06	2011
10	J CLIN NURS	75	0.07	2013

In summary, the journal co-citation network reveals a dual structure: specialized journals such as HERD anchor the field, while high-impact medical and interdisciplinary journals provide essential cross-disciplinary connections. Together, they illustrate the hybrid nature of hospital design research, bridging architecture, healthcare, and behavioral sciences.

4.3. Knowledge Base and Disciplinary Structure Evolution

4.3.1. Analysis of the Cited References

Reference co-citation analysis, first proposed by Small [128], is based on the principle that if two papers are cited together by a third, they share a co-citation relationship. This method helps identify the intellectual foundations of a field and trace its evolving research frontiers.

The cited reference co-citation network generated by CiteSpace (version 6.2.R3) (Figure 9) includes 966 nodes and 2,305 links, with a network density of 0.0049. This indicates that citation connections exist but remain relatively scattered, reflecting the diversity of scholarly contributions in hospital design.

The network presents a clear dual-core structure and suggests a consolidation of evidence-based design as the theoretical backbone, while peripheral applied clusters reflect evolving practical interventions. At its center are seminal works on evidence-based design (EBD), particularly Ulrich R.S. (2004) [41] and Ulrich R.S. (2008) [39]. These references have both a high citation frequency and high Betweenness Centrality, establishing them as the foundational literature in the hospital design knowledge system.

On the other hand, more recent publications, such as Braun V. (2021) [129], Taylor E. (2018) [130], and Brambilla A. (2019) [12], form tightly connected sub-clusters. These emerging studies reflect the field’s shift from theoretical foundations toward practical interventions and environmental experience, particularly in response to challenges before and after the COVID-19 pandemic.

In summary, the co-citation network reveals a field anchored by the seminal EBD literature while simultaneously expanding through new clusters of applied research. This dual pattern highlights both the stability of core theoretical foundations and the dynamic evolution of hospital design research in recent years.

The co-citation clustering network (Figure 10) displays a typical “center–periphery” structure. At the core, the evidence-based design (EBD) literature anchors the network, while peripheral clusters capture topics of applied practice and methodological development. Together, they outline the knowledge spectrum of hospital design, spanning from conceptual foundations to functional applications.

The clustering quality is robust, with Q = 0.8076, indicating a strong modular structure, and S = 0.9206, reflecting high internal consistency and clear differentiation among clusters. A total of nine major knowledge clusters were identified between 1932 and 2025:

• Cluster #0 (evidence-based design): This cluster focuses on optimizing hospital design based on empirical research, emphasizing the impact of the physical environment on patient recovery and employee performance. As the core and most node-dense area of the entire network, it shows that EBD is the most representative and structurally stable theoretical foundation in this field.
• Cluster #1 (realistic evaluation) and Cluster #3 (healthcare facilities): They focus on “design effect evaluation” and “medical institution design practice”, respectively, explore the methodology and practical application challenges of the effect evaluation of healthcare facilities environmental intervention measures, as well as the functional layout optimization and design evolution, forming relatively independent but interconnected secondary research topics in the network structure.
• Cluster #2 (engineering infection control) and Cluster #8 (single- versus multiple-occupancy room): Located at the edge of the network but with clear cluster boundaries, they focus on how architectural engineering design intervenes in hospital infection prevention and control, especially in ventilation systems and spatial layout strategies. They also compare single-patient rooms to multi-occupancy rooms in terms of infection control, patient privacy, and treatment effects. This reflects the importance of research on infection control and ward type selection at a specific stage.
• Clusters #4 (sustainable finishing material) and #6 (environmental criteria): Research on the application of sustainable materials in medical buildings, emphasizing environmental performance and indoor health standards and focusing on environmental performance standards for healthcare environment design, including air quality, energy consumption, lighting conditions, etc. They show the expansion trend of green medical building research in recent years.
• Cluster #5 (positive distraction): As an interdisciplinary integration attempt, it reflects the research related to “psychological healing design” in the medical space and analyzes the strategy of providing positive sensory stimulation (such as natural landscape and color application) through architectural design to promote patient recovery.
• Cluster #7 (research literature): It covers review citations of the existing research and summarizes and generalizes important research methods in the field of hospital design, indicating that this topic is mostly directed towards methodology and literature system construction.

In summary, the clustering analysis highlights EBD as the theoretical core of hospital design, while newer clusters reflect applied themes such as infection control, sustainability, and healing environments. Together, these clusters demonstrate the field’s dynamic balance between stable conceptual foundations and expanding interdisciplinary applications.

The co-citation clustering timeline view (Figure 11) illustrates the active periods and duration of each thematic cluster, providing a temporal perspective on the evolution of hospital design research.

Before 2005, only a few clusters, such as Cluster #2, Cluster #7, and Cluster #8, were active. These topics reflect the dominance of basic theories and methodological foundations, including infection control, evaluation methods, and environmental effects. Between 2005 and 2015, new clusters gradually emerged and expanded. Cluster #0 and Cluster #5 became increasingly prominent, highlighting a growing focus on grounded theory, patient-centered approaches, and psychological support in healthcare architecture. Since 2016, research priorities have shifted again. Clusters #1, #3, #4, and #6 gained prominence, reflecting the field’s movement toward green design, humanistic care, and resilient responses to disease outbreaks.

In summary, the timeline view demonstrates the ongoing transformation of hospital design research. Early emphasis on infection control and assessment methods has gradually given way to themes of sustainability, resilience, and patient-centered healing, marking the continuous evolution and expansion of the field’s knowledge structure.

Beyond identifying well-established thematic clusters, the analysis also reveals several under-researched areas that merit further investigation. For instance, psychiatric care environments, maternity facilities, and post-disaster or temporary hospitals are seldom represented in the co-citation clusters despite their clear practical importance. Likewise, clusters addressing sustainable materials and environmental criteria remain peripheral, suggesting that the integration of green and low-carbon design principles is still at an early stage. Furthermore, the limited presence of information technology–related themes indicates that digital healthcare infrastructure and smart hospital design have yet to be systematically incorporated into the knowledge base. These gaps highlight promising directions for future research, including broadening thematic coverage to underrepresented facility types, strengthening sustainable and low-carbon design approaches, and embedding digital and data-driven perspectives into hospital design research.

4.3.2. Analysis of the Subject Categories

The scope of a subject and its development trajectory can be explored by studying its disciplinary classification [131]. Moreover, the importance and applicability of each subject can be determined, highlighting the correlation between disciplines and indicating potential research directions [132].

Table 5. Top 10 major subject categories. (Note: “Year” refers to the first year in which the WoS category appeared in the retrieved dataset).

No.	WoS Categories	Count	Centrality	Year
1	PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH	203	0.44	1958
2	NURSING	91	0	2007
3	MEDICINE, GENERAL & INTERNAL	84	0	1932
4	HEALTH CARE SCIENCES & SERVICES	59	0.58	1960
5	CONSTRUCTION & BUILDING TECHNOLOGY	53	0.17	1983
6	INFECTIOUS DISEASES	46	0.12	1981
7	HEALTH POLICY & SERVICES	39	0.05	1965
8	ENGINEERING, CIVIL	38	0.05	2011
9	PSYCHIATRY	29	0.18	1948
10	ARCHITECTURE	27	0.05	1975

lists the number of documents, the Betweenness Centrality, and the first appearance time of each major subject. It can be seen that among the top ten high-frequency subjects, “PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH” has the highest frequency. It occupies a hub position in the entire subject category co-occurrence network and plays a bridging role between multiple subjects, showing that it plays a core hub role in interdisciplinary research. “HEALTH CARE SCIENCES & SERVICES” ranks fourth, with the highest centrality value (Betweenness Centrality = 0.58), which connects medicine, policy, nursing, and architecture subjects. It is noteworthy that “NURSING” and “MEDICINE, GENERAL & INTERNAL” show a centrality value of zero. This does not indicate a lack of relevance but rather that these categories do not serve as bridges in the disciplinary co-occurrence network. They are positioned as specialized or peripheral nodes that contribute within their clusters but do not appear on the shortest paths linking multiple clusters. Such categories may thus represent specialized or more peripheral disciplinary contributions, which play localized rather than intermediary roles in the knowledge structure of hospital design research.

The subject category co-occurrence network (Figure 12) shows that the hospital design field has obvious multidisciplinary characteristics. Except for “PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH”, traditional medicine-related subjects such as “NURSING”, “MEDICINE, GENERAL & INTERNAL”, “HEALTH CARE SCIENCES & SERVICES”, and “CONSTRUCTION & BUILDING TECHNOLOGY” form the main research basis of this field and are in the core position of the network, indicating that this field has always been based on medical health. “ARCHITECTURE” has a small node, but it has a direct connection with “CONSTRUCTION & BUILDING TECHNOLOGY” and “PUBLIC HEALTH”, showing that architectural knowledge has been absorbed and applied to hospital design research. The small but strongly linked ‘Architecture’ node highlights disciplinary asymmetries, suggesting that architectural scholarship is underrepresented in publication counts but is continuously integrated through interdisciplinary exchange. In addition, there is a strong co-occurrence relationship between subjects such as “ENVIRONMENTAL SCIENCES”, “ENGINEERING, CIVIL”, and “HEALTH POLICY & SERVICES”.

The subject category time zone view in Figure 13 further reveals the activity and evolution path of different subject categories on the research timeline. The earliest category to appear was “MEDICINE, GENERAL & INTERNAL” (1932), followed by “PUBLIC HEALTH”, “PSYCHIATRY”, “ARCHITECTURE”, etc., meaning that the research field was initially dominated by medicine and gradually expanded to the environment, architecture, and policy, showing a trend from a single medical orientation to multidisciplinary integration. After entering the 21st century, emerging categories such as “MEDICAL INFORMATICS”, “COMPUTER SCIENCE, INFORMATION SYSTEMS”, and “ENGINEERING, MULTIDISCIPLINARY” began to appear, showing a trend that the research content has expanded towards informatization and technology integration in recent years.

In summary, subject category analysis demonstrates that hospital design research originated in medicine, progressively absorbed insights from public health and architecture, and is now entering a new stage of technological and interdisciplinary integration.

4.4. Research Topics, Hotspots, and Future Trends—Analysis of the Keywords

Keywords can reflect the main content, style, and core concepts of an article [133]. However, it should be noted that many older WoSCC records do not include author keywords. This absence may lead to an underrepresentation of early research themes and partly explain abnormal growth patterns in keyword-based analyses [134]. On this basis, the more links there are, the more frequently the two keywords appear together; the larger the nodes are, the deeper the correlation and association between them; and the thicker the connection line is, the stronger the relationship between them [135]. CiteSpace (version 6.2.R3) calculates the co-occurrence frequency and Betweenness Centrality of terms and generates a keyword co-occurrence network (Figure 14) to demonstrate the popular trends and knowledge structure effectively [136].

The node “hospital design”, as a foundational term in the research field, is high in frequency and centrality, connects to multiple other sub-themes, and is the largest and most central node. The “healthcare” and “impact” nodes are also high in frequency and centrality, indicating that the impact of hospital delivery systems and health outcomes is a central concern of the study. The appearance of the nodes “built environment”, “EBD”, and “patient safety” shows a gradual expansion of research from physical design to user experience, health performance, and safety dimensions. In addition, keywords such as “management”, ‘satisfaction’, “communication”, “quality”, and “model” are found around the center, forming multiple thematic branches, reflecting the multidimensional cross-cutting structure of hospital design research, covering design methodology, evaluation criteria, behavioral psychology, operational efficiency, etc.

In summary, the keyword co-occurrence network highlights a shift from core architectural concerns toward integrated perspectives that link design with health outcomes, patient experience, and system performance, demonstrating the increasingly interdisciplinary nature of hospital design research.

By analyzing the high-frequency changes of each keyword in a certain period, further exploring the main content of the key literature in the research field becomes possible, thereby revealing the transformation of popular research topics and research patterns [137]. The burst of keywords identified 28 keywords with significant burst strength (

Table 6. Top 28 keywords with the strongest citation bursts. (Note: The horizontal bar represents the timeline from 1932 to 2025. The deep blue segment indicates the period when the keyword actually appeared in the dataset. The light blue segment represents the remaining time span within the study period without keyword occurrence. The red segment highlights the burst period, when the keyword received extraordinary attention and citations).

Keyword	Year	Strength	Beginning	End	1932–2025
intensive care unit	1998	3.14	1998	2010
evidence-based design	2007	3.82	2007	2013
satisfaction	2012	5.44	2012	2017
people	2021	3.69	2021	2025
patient safety	2007	3.43	2015	2019
management	2001	2.84	2017	2021
work	2021	2.36	2021	2025
patient	2018	3.79	2022	2025
quality	2016	3.1	2018	2021
healthcare	2001	2.28	2007	2010
mental health	2021	4.06	2021	2023
therapeutic landscapes	2013	3.7	2013	2015
performance	2020	3.48	2020	2022
space	2019	3.26	2019	2021
model	2012	2.91	2020	2022
safety	2020	2.73	2020	2022
mortality	2018	2.55	2020	2022
impact	2001	5.78	2022	2023
hospital design	1991	3.92	2008	2009
environmental design	2021	3.27	2021	2022
design process	2013	3.22	2013	2014
surgery	2018	2.79	2018	2019
environments	2013	2.7	2022	2023
experience	1994	2.51	2020	2021
outcome	2012	2.4	2021	2022
risk	2010	2.37	2018	2019
stress	2014	2.34	2021	2022
framework	2021	2.3	2024	2025

).

The three keywords with the highest burst strength are “impact” (Strength = 5.78, 2022–2023), “satisfaction” (Strength = 5.44, 2012–2017), and “mental health” (Strength = 4.06, 2021–2023). These terms appear frequently in the core literature within concentrated timeframes, reflecting periods of intensive scholarly attention. “Impact” and “satisfaction” emphasize the assessment of healthcare facilities on user outcomes and satisfaction, showing a shift from spatial design to performance evaluation. “Mental health” highlights the growing importance of psychological healing environments, particularly in the post-pandemic era.

The keywords with the longest burst duration include “intensive care unit” (1998–2010) and “evidence-based design (EBD)” (2007–2013). Their persistence underscores sustained research value. “Intensive care unit” reflects the longstanding concern with functionally sensitive spaces, while EBD remains a foundational methodology for evaluating the effectiveness of medical environments.

In terms of newly emerging bursts, keywords such as “people” (2021–2025), “framework” (2024–2025), “patient”, “performance”, “safety”, and “environmental design” represent likely research frontiers. These terms signal the rise of people-centered design logic, the pursuit of systematic framework development, and a stronger emphasis on performance-oriented evaluation. Notably, “framework” is a novel burst, occurring since 2024, reflecting exploratory interest in concepts such as “universal design frameworks” and “cross-scale design strategies,” although its intensity remains modest.

In summary, keyword burst analysis reveals a clear trajectory of hotspot migration: from functional layout and safety assurance to the widespread adoption of EBD and, more recently, to themes of user experience, mental health, and environmental integration. This indicates that hospital design research is increasingly converging with contemporary social and health needs.

The keyword clustering network (Figure 15) was generated using CiteSpace’s Log-Likelihood Ratio (LLR) algorithm, which automatically groups co-occurring keywords into thematic clusters. This method enables the identification of underlying research structures and the calculation of quality metrics to assess clustering validity. The network presents a clear modular structure, indicating that the research themes are relatively independent with well-defined boundaries.

The clustering quality metrics confirm the robustness of the results: Q = 0.8076, suggesting a strong modular division effect, and S = 0.9206, indicating high internal consistency and reliable differentiation. A total of 11 major clusters (Clusters #0–10) were identified, each reflecting distinct thematic areas in hospital design research.

• Cluster #0 and Cluster #1 are the largest, reflecting their significant aggregation and high research density.
• Cluster #2 highlights terms such as “hospital design”, “physical environment”, and “treatment context”, emphasizing the growing importance of patient-centered design and experiential outcomes.
• Cluster #4 includes terms like “energy efficiency” and “pollution prevention”, signaling the rise of green hospitals and sustainable building practices.
• Cluster #10 emphasizes “biophilic design parameters” and “natural material”, illustrating the trend of integrating natural elements into healthcare spaces.
• Smaller clusters capture more specialized themes, such as population-specific psychological healing and the design of psychiatric hospital environments, pointing to the diversification of research interests.

In summary, the clustering results show that the current knowledge base in hospital design is composed of multiple parallel themes: traditional hospital design, patient-centered spatial strategies, sustainable and green building practices, biophilic and nature-integrated approaches, and interdisciplinary interventions. This reflects the continuous expansion and diversification of the field.

The keyword clustering timeline view (Figure 16) illustrates the temporal distribution and persistence of research themes, highlighting the evolutionary trajectory of hospital design studies. The horizontal axis represents the years (1991–2025), while the vertical axis corresponds to cluster labels. Each cluster’s duration and intensity reflect its emergence, development, and decline.

In the early 2000s, clusters such as Cluster #0 and Cluster #4 were active, focusing on spatial configuration and specific treatment contexts (e.g., addiction treatment). These clusters reveal the initial emphasis of healthcare building design on the functional layout and treatment-specific environments.

Between 2005 and 2015, Cluster #5 and Cluster #6 became increasingly prominent, reflecting deeper exploration of emergency processes and specialized hospital areas. During this period, Cluster #3 remained active, with keywords centered on ICU design and its relationship to clinical outcomes, underscoring the growing recognition of critical care environments as a research priority.

Since 2015, new clusters, such as Cluster #2, Cluster #8, and Cluster #10, have emerged and gained momentum. These clusters signify a paradigm shift from safety- and function-oriented design to approaches emphasizing patient experience, health promotion, sustainability, and environmental integration.

In summary, the timeline view demonstrates how hospital design research has evolved from functional layouts and safety concerns to interdisciplinary paradigms prioritizing patient well-being and environmental responsiveness, reflecting the field’s dynamic adaptation to societal and healthcare needs.

4.5. Summary of Results

The main findings of Section 4are summarized and presented in

Table 7. Findings and interpretations.

Analysis Dimension	Key Findings	Interpretation
Publication Trends	Total of 877 papers (1932–2025); rapid growth since 2020 (417 papers; 47.56% of total).	Reflects both WoSCC coverage expansion and a genuine surge of interest post-COVID-19; indicates hospital design has become a rapidly developing interdisciplinary field.
Country Collaboration	The USA leads (260 papers, high centrality), followed by the UK and Australia; China and Canada are emerging; Japan, Turkey, and Italy act as regional bridges.	Global research is concentrated in developed countries; some nations act as strategic connectors. Regional disparities persist (e.g., Africa and South America are underrepresented).
Institution Collaboration	Leading institutions: Univ. of London, Aarhus Univ., and Univ. of Toronto; regional clusters in Europe, North America, and Asia-Pacific.	Institutional collaboration remains fragmented, with geographic clustering; scope for broader international partnerships.
Author Collaboration	713 authors, 767 links, network density of 0.003; mostly small groups, with weak cross-group links. Core figures include Ulrich RS, Joseph A, and Chaudhury H.	The field lacks strong central leaders; collaboration is mostly localized. Influential authors provide theoretical and methodological anchors.
Highly Cited References	13 papers with >100 citations (2004–2012 peak). Core works: Ulrich (2004, 2008).	Established EBD as the theoretical foundation; research gradually shifted toward patient-centered and infection control design.
Author Co-Citation Network	Key authors: Ulrich RS (central node), Joseph A, Chaudhury H, Hamilton DK, and Pati D.	Indicates strong reliance on EBD pioneers and healthcare design scholars, forming the intellectual base of the field.
Journal Co-Citation Network	Core journals: HERD, Environment and Behavior, The Lancet, NEJM, and JAMA.	The field is highly interdisciplinary, combining architecture, medicine, psychology, and nursing.
Subject Categories	Core: Public health, nursing, medicine, construction technology, and architecture. Emerging: Medical informatics and computer science.	The field is evolving from medical dominance to multidisciplinary integration, including digital and environmental sciences.
Keyword Analysis	Hotspots: “hospital design”, “healthcare”, “impact”, “EBD”, and “patient safety”. Emerging: “mental health”, “framework”, and “environmental design”.	Research focus has shifted from space/function to user-centered outcomes, sustainability, and psychosocial healing; it indicates future emphasis on resilience and system-level frameworks.

. The table provides an overview of key observations across multiple dimensions of bibliometric analysis. It clearly separates objective findings (e.g., publication counts, centrality values, and clustering outcomes) from interpretive implications (e.g., global disparities, theoretical foundations, and emerging research frontiers). This structured synthesis complements the detailed narratives in Section 4. Furthermore, it sets the stage for a deeper discussion in the following section.

5. Discussion

This study systematically mapped the knowledge structure and developmental trajectory of hospital design by employing cooperation networks, co-citation networks, disciplinary co-occurrence analysis, and keyword evolution analysis. These methods reveal the core characteristics of the field over its long-term evolution as well as the accelerated research momentum following the COVID-19 pandemic.

This study found that since 2020, the volume of relevant works has increased markedly. While the United States continues to hold a dominant position, countries such as China, Turkey, and Italy are emerging, creating a more diversified landscape of academic participation. Nevertheless, author and institutional collaboration networks remain fragmented, indicating considerable room for strengthening international cooperation and interdisciplinary integration. Unlike existing studies that focus on individual cases or specific topics, this work provides a comprehensive examination of the paradigm shift from evidence-based design (EBD) to patient-centered Environmental Healing Design (EHD), drawing on nearly a century of literature. Through visualization, it also uncovers the dynamic links between knowledge structures and societal health needs—an aspect that has been less systematically addressed in previous research. Furthermore, architecture is increasingly converging with fields such as public health, environmental science, and information technology, reflecting a growing trend toward interdisciplinary integration. However, a unified evaluation framework is still lacking, and theoretical as well as methodological coherence across disciplines remains limited. The recent emergence of “framework” as a keyword may signal rising scholarly attention to this issue.

Importantly, this bibliometric analysis also provides empirical evidence addressing the gaps identified in the literature review. First, by mapping global collaboration networks, it quantitatively demonstrates the persistent dominance of Western countries while highlighting the emerging roles of China, Turkey, and Italy, thus offering an updated evidence base for addressing regional disparities. Second, although psychiatric hospitals and temporary facilities remain underrepresented as subfields, keyword clustering and burst analyses reveal growing interest in mental health and emergency response design, suggesting a gradual shift toward more diverse facility types. Third, co-citation and subject category analyses confirm the increasing involvement of disciplines such as public health, environmental science, and information technology, indicating that the gap in interdisciplinary integration is beginning to narrow. Collectively, these findings show that bibliometric analysis not only identifies but also empirically traces the evolution of these gaps, thereby providing a stronger foundation for future targeted research.

This study has made some contributions to the theoretical organization and graphical analysis, but there are still limitations. First, the dataset was retrieved exclusively from the Web of Science Core Collection (WoSCC). Although the WoSCC provides standardized citation metadata essential for co-citation and co-occurrence analyses, it may not fully cover local or non-English publications. Both the WoSCC and Scopus have been shown to underrepresent journals from certain regions, which may bias the global visibility of non-Western scholarship [106,107]. This coverage bias introduces uncertainty in the global comparability of results, as the dominance of Western countries may partly be overstated [138].

Second, although the WoSCC nominally covers records since 1900, bibliometric retrieval of the older literature is subject to coverage limitations, as many early records lack complete metadata [134]. Consequently, the relatively sparse output before the 1950s may partly reflect database gaps rather than a true absence of scholarly activity. This underrepresentation may lead to an underestimation of the historical research activity in hospital design. Missing metadata, such as incomplete author address information, has also been reported to distort long-term collaboration analysis [111].

Third, WoSCC coverage has expanded over time through the inclusion of new datasets, which may artificially inflate publication counts in certain periods [108,109]. In our case, while the sharp increase in research output after 2020 reflects genuine academic attention driven by COVID-19 and public health challenges, it may also have been amplified by WoSCC coverage growth.

Finally, there are methodological limitations associated with the use of CiteSpace. CiteSpace relies primarily on co-citation and co-occurrence frequencies, which are structurally oriented and may overlook semantic and contextual nuances. This means that the identified clusters and knowledge structures may simplify complex interdisciplinary linkages, thereby introducing some uncertainty in interpreting the intellectual structure of the field [103]. Expanding the dataset to include other databases (e.g., Scopus, PubMed, CNKI, and KCI) could also improve coverage and enhance the diversity of perspectives included in future bibliometric studies [139]. Moreover, the WoSCC is a dynamic database that is continuously updated, which means that replication at a different time may yield slightly different results. It is strongly biased toward English-language journals and uses subject categories that are not always precise and may overlap [140]. While these factors may introduce uncertainty into specific results, we believe they do not undermine the robustness of the overall trends and knowledge structures identified in this study.

In summary, this study delineates the knowledge base, collaboration patterns, and evolutionary pathways of hospital design research from a multidimensional perspective. By illustrating how the field is transforming amid global crises and disciplinary convergence, it provides a theoretical foundation and practical reference for developing more resilient, inclusive, and evidence-based healthcare environments. Despite the limitations noted above, we maintain that the overall trends and knowledge structures identified here are robust and offer valuable insights into the evolution of hospital design research.

6. Conclusions

Drawing on the literature from 1932 to 2025 retrieved from the WoSCC, this study conducted a systematic bibliometric analysis of hospital design research across publication trends, collaboration networks, co-citation structures, disciplinary systems, and research hotspots, using tools such as CiteSpace (version 6.2.R3), Excel, and Tableau (version 2024.3).

The results reveal a clear trajectory: early studies were fragmented and function-oriented, whereas contemporary research reflects a paradigm shift toward patient-centered and evidence-based design. Since 2020, the field has entered a rapid growth phase, characterized by a strong emphasis on resilience, psychological well-being, and environmental integration. This transformation illustrates the evolution of hospital design from a narrow architectural concern to a multidimensional domain that actively responds to global health crises and fosters interdisciplinary knowledge exchange.

The findings also underscore persistent structural challenges. Collaboration networks remain fragmented, with limited cross-regional and cross-disciplinary integration, particularly between the Global North and South. Architectural scholarship, while central to the field, appears underrepresented compared to contributions from medicine and public health, highlighting the need for a more balanced disciplinary structure. The dominance of a few core authors and institutions reinforces the influence of established centers but raises concerns about inclusivity and diversity in shaping global knowledge production.

As noted in the Discussion, limitations exist, including reliance on a single database and incomplete metadata in older records. While these factors may affect coverage, they do not undermine the robustness of the long-term trends identified.

Future research should address these gaps in several ways. First, expanding database coverage and incorporating multilingual and regional journals can mitigate language and geographic biases. Second, methodological innovation—such as full-text mining, semantic analysis, and mixed-method integration—can enrich bibliometric findings with qualitative insights. Third, linking bibliometric evidence to post-occupancy evaluations and health outcome studies can move hospital design research closer to policy and practice, ensuring that theoretical advances translate into measurable improvements in healthcare environments.

In conclusion, this study clarifies the knowledge base, core contributors, and evolutionary pathways of hospital design while underscoring its dynamic responsiveness to societal challenges. By identifying both achievements and gaps, the findings provide a roadmap for advancing toward more resilient, adaptive, and human-centered healthcare environments, offering both theoretical and practical guidance for the next stage of research and practice. For scholars, strengthening interdisciplinary collaboration among architecture, medicine, psychology, and information science will be essential. For practitioners, adopting evidence-based and patient-centered design strategies—such as healing environments, flexible layouts, and sustainable materials—can improve design quality and health outcomes. For policymakers, supporting inclusive research funding, promoting wider dissemination of regional and non-English studies, and developing standards that enhance resilience and equity in healthcare facilities are critical steps. Together, these targeted actions can help shape a more resilient, adaptive, and human-centered healthcare future.