A Review of Research Progress on the Impact of Urban Street Environments on Physical Activity: A Comparison between China and Developed Countries

Abstract

Creating healthy street environments to encourage physical activity is an effective strategy against non-communicable diseases exacerbated by rapid urbanization globally. Developing countries face more significant health challenges than developed ones. However, existing research predominantly focuses on the perspective of developed countries. To address the health challenges in developing nations, studies should not only draw on the findings from developed countries but also clearly define unique research processes and pathways. Consequently, this study conducts a comparative analysis between China, representing developing countries, and developed nations, using databases like China National Knowledge Infrastructure (CNKI) and Web of Science (WOS) and tools such as Citespace, Bicomb, and Statistical Package for the Social Sciences (SPSS) to explore research hotspots, developmental trajectories, thematic categories, and trends. The findings reveal a shift in developed countries from macro-material to micro-environmental elements under multidisciplinary scrutiny, while future topics may include street space evaluations and psychological healing. In China, research has been dominated by different disciplines at various stages, starting with medical attention to chronic disease prevention, which then shifted to traffic engineering’s focus on constructing green travel environments, and finally expanded to disciplines like landscape architecture examining the impact of street environment elements on pedestrian behavioural perceptions. Future themes will focus on promoting elderly health and urban health transport systems. Generally, research in developed countries exhibits a “bottom-up” approach, with practical issues at a “post-evaluation” stage, primarily based on the “socio-ecological model” and emphasizing multidisciplinary collaboration. Chinese research shows a “top-down” characteristic, driven by national policies and at a “pre-planning” stage, integrating theories such as Maslow’s hierarchy of needs and attention restoration theory, with relatively loose disciplinary cooperation. Overall, research is shifting from macro to human-centric scales and is progressively utilizing multi-source and multi-scale big data analysis methods. Based on this, future research and development recommendations are proposed for developing countries, with China as a representative example.

1. Introduction

The coronavirus disease 2019 (COVID-19) pandemic has severely impacted global health, hindering progress toward the Sustainable Development Goals [1]. In 2023, the first Healthy City Partnership Summit in London brought together mayors and officials from over fifty major cities. The summit highlighted the renewed focus on non-communicable diseases and injuries three years after the COVID-19 pandemic outbreak, noting that cities have unique advantages in combating non-communicable diseases and reducing injuries by implementing policies that significantly reduce exposure to risk factors, thus helping the world return to the trajectory of the Sustainable Development Goals [2]. The global urbanization rate is expected to rise to 68% by 2050 [3], with an additional 2.5 billion urban inhabitants. Most of this urban population growth will occur in developing countries, especially in China, India, and Nigeria [4]. Issues such as poor sanitation, air pollution, road safety, and limited access to healthy food and spaces during urbanization elevate the risk of non-communicable diseases [5]. Non-communicable diseases cause 41 million deaths annually, accounting for 74% of all deaths globally, with 77% occurring in low- and middle-income countries [6]. These countries face more severe health challenges than developed nations due to several factors: a lack of quality healthcare services, inadequate social welfare systems, insufficient dissemination of disease prevention knowledge, and the rapid progression of urbanization and aging populations [7]. The World Health Organization (WHO) has identified urbanization as one of the leading public health challenges of the 21st century [8]. Evidence-based medical research has proven that a lack of physical activity is one of the top ten global death risk factors [9] and increasing physical activity and reducing sedentary behaviour could prevent at least 3.2 million deaths related to non-communicable diseases globally each year [10]. WHO has emphasized that urban planning should actively contribute to addressing health challenges, highlighting that “health must be a primary focus for urban planners” [11]. Streets, as a crucial component of urban spaces, can directly influence human health by providing residents with comfortable venues for physical activity [12,13]. They can also indirectly promote health by reducing factors that potentially impact traffic accidents, thereby ensuring the safety of residents’ activities [14,15]. Health-oriented street design is an effective way to foster the creation of healthy cities [16].

As the world’s largest developing country, over the past 40 years, China has seen its urbanization rate increase from 17.92% to 65.2% [17], which exceeds significantly that of other countries during the same period, making it one of the fastest urbanizing countries globally [18]. According to United Nations projections, China’s urbanization rate will reach 74% by 2035, meaning that, in the next decade, over 100 million rural inhabitants will move to urban areas [19], resulting in a vast urban population. Like other developing countries, China faces health challenges due to shifts in disease patterns and an aging population. Non-communicable diseases have become the most significant health threat to Chinese citizens, accounting for approximately 88.5% of all deaths [20]. Additionally, due to the rapid development of the automotive industry, the physical environment of streets, including air quality and urban transportation infrastructure, complicates the risk of incidence and disease prevention efforts. The task of building healthy cities in China is unprecedented in its enormity and complexity. To address this challenge, the Chinese government has steadfastly implemented the policy of “integrating health into all policies” to promote the implementation of health governance measures. In 2016, the Chinese State Council issued the “Healthy China 2030” blueprint outline, placing health as a strategic priority in national development. This plan significantly measures China’s active participation in global health governance and fulfilling its commitments to the United Nations’ “2030 Agenda for Sustainable Development” [21]. In 2018, China established the “National Health City Evaluation Indicator System” and began health city evaluations based on this system [22]. In 2019, the “Opinions of the State Council on Implementing the Healthy China Action” were promulgated, which specified measures in the area of national fitness and health to prevent and control road traffic injuries. Other related documents have also proposed strengthening the construction of urban greenways, fitness trails, and other health-oriented living environments [23]. In 2021, the State Council issued the “National Fitness Plan (2021–2025)”, which set forth requirements for enhancing the public service system for national fitness and making sports and fitness activities more accessible to the public [24]. This reflects the emphasis, on a national scale, on creating a healthy urban environment that promotes residents’ physical activity.

Physical activity (PA) refers to any bodily movement produced by skeletal muscles that requires energy expenditure. The urban street environment includes the physical environment (natural and artificially modified built environment) and social environment (such as social norms and community cohesion). It is an essential factor that affects people’s physical activity and is also an important entry point for cities to intervene in people’s health proactively. Scholars from various fields, including public health and preventive medicine [25], ecology [26], urban planning [27], and sports science [28], have extensively and deeply researched the impacts of land mix use [29], street connectivity [30], walkability [27], and the built environment [31] on physical activity. However, current research is primarily concentrated in developed countries, such as the United States (which accounts for 44.1% of publications) and Australia (11.9% of publications). Among developing countries, China produces the most research in this field, with its output second only to the United States (16.9% of publications). As these research interests span multiple disciplines, the related literature is extensive and dispersed across various journals. Considering the differences in development stages and planning governance methods between countries, the practical directions, pathways, and depth of research on the impacts of street environments on physical activity vary, yet discussions on this topic have not been previously synthesized. Additionally, existing review articles mainly organize and evaluate research on the impact of urban architectural environments and green spaces on physical activity among different age groups. For instance, Zhang et al. used a narrative review method to summarize existing reviews on the association between architectural environments and physical activity among children, adults, and the elderly [32]. Prince et al. systematically described the current status, intensity, and quality of research on the association between the built environment and adult physical activity in high-income countries [33]. Ran et al. used Citespace and VOSviewer software for a bibliometric analysis of the association between urban green space features and the occurrence of leisure physical activity [34].

In conclusion, compared to developed countries, developing nations face more pronounced health challenges. Current research on the “impact of street environments on physical activity” is primarily conducted from the perspective of developed countries. However, due to differences in political systems, socioeconomic environments, urban infrastructure, and demographic structures between developing and developed countries, the research findings based on developed countries’ contexts are not entirely applicable to developing nations. To address the health challenges faced by developing countries, it is necessary to both draw on the research outcomes from developed countries and to clearly define national research agendas and approaches, thereby more actively addressing the foreseeable health challenges in the process of urbanization. Therefore, the study selects China as representative of developing countries to conduct comparative research with developed countries. Using a systematic review methodology and applying software tools such as Citespace (version 6.2.R6), Bibliographic Items Co-occurrence Matrix Builder (Bicomb) 2.0, and Statistical Package for the Social Sciences (SPSS) 27.0, the study performs a systematic summary of research hotspots for studies in China and developed countries, organizes the development trajectories and thematic categories, and scientifically predicts research trends. This analysis deeply explores the differences between research in China and developed countries and the reasons for these differences (as shown in Figure 1). The findings provide a reliable basis and inspiration for future research by Chinese scholars, helping international scholars to deeply understand the current state of research on the impact of street environments on physical activity within the Chinese context.

Accordingly, the structure of this paper is as follows: Section 2 provides an overview of the research materials and methods; Section 3 presents the research hotspots, development contexts, thematic categories, and trends in China and developed countries; Section 4 delves into the similarities and differences between the research in China and developed countries, explores the reasons for these differences, and proposes future directions for research in China; Section 5 concludes the paper.

2. Research Methods

2.1. Literature Selection

The study separately conducted literature searches on the impact of street environments on physical activity in developed countries and China. For English-language literature, the search was conducted using the Web of Science (WOS) core database with the search formula ((TS = (street)) AND TS = (physical activity)) AND PY = (1996–2023). The search included the Science Citation Index-Expanded (SCI-E), the Social Sciences Citation Index (SSCI), and the Arts & Humanities Citation Index (A&HCI), which were limited to English-language literature. Subsequently, by excluding literature irrelevant to the study of street environments’ impact on physical activity based on titles and abstracts, 1484 entries were retrieved. For Chinese-language literature, the China National Knowledge Infrastructure (CNKI) database was utilized, which is the largest continuously updated database of Chinese academic literature [35], effectively supplementing the English database and providing a comprehensive view of research in China. To ensure the quality of the papers, journals indexed by Science Citation Index (SCI) or Engineering Index (EI) and core journals from Peking University, Chinese Social Sciences Citation Index (CSSCI) journals, and Chinese Science Citation Database (CSCD) journals were selected, covering the period from 1993 to 2023. The search terms used were “street” AND “physical activity”, initially yielding 17 papers. Further searches with expanded thematic keywords, such as “urban” AND “physical activity”, “street” AND “slow traffic”, and “urban” AND “slow traffic”, were conducted, and titles and abstracts were again screened to exclude irrelevant studies, ultimately resulting in 700 papers.

2.2. Research Steps

Firstly, data handling was carried out, wherein the literature data from WOS and CNKI were preprocessed using the Citespace (version 6.2.R6) built-in data converter. To ensure accuracy, after the format conversion, data from WOS and CNKI were filtered to remove duplicates, with 1434 English-language papers and 665 Chinese-language papers determined to be valid. In the second step, research hotspots were identified. The filtered data were imported into Citespace (version 6.2.R6) for bibliometric analysis, setting the time slice to one year and selecting “Keyword” as the node type, and keyword co-occurrence analysis was conducted separately for international and Chinese research fields related to the impact of street environments on physical activity. Furthermore, development trajectories were constructed. Both English- and Chinese-language literature annual publication volumes were visualized using Excel. Based on this, along with the keyword time zone distribution map and keyword prominence map generated by Citespace (version 6.2.R6), the development trajectories of research on the impact of street environments on physical activity in international and Chinese contexts were established. With regard to fourth step, a keyword clustering analysis was undertaken. Using Bicomb2.0, the data from both English- and Chinese-language literature were processed to identify high-frequency keywords and generate term–document matrices. After processing the term–document matrix data in Excel, they were imported into SPSS 27.0 software, where hierarchical clustering analysis was conducted to obtain dendrograms, followed by keyword clustering analysis. Finally, research trends were interpreted. By utilizing the multidimensional scaling feature in SPSS 27.0 software, multidimensional scaling analysis maps were constructed. These maps facilitate a scientific interpretation of research trends based on the distribution quadrants of keywords.

3. Research Progress and Trends on the Impact of Street Environment on Physical Activity in China and Developed Countries

Keywords play a crucial role in succinctly representing a paper’s core content and important concepts [36]. The literature can be transformed into quantifiable analytical data by analysing information such as keyword co-occurrence, clustering, temporal distribution, and emergent trends. Particularly, the refinement of high-frequency keywords can, to some extent, reflect scholars’ attention to specific topics, characterizing the hotspots and frontier trends in the research field.

3.1. Research Hotspots

The co-word analysis of keywords can reveal the hotspots and internal connections of specific research topics. The basic principle involves counting the occurrences of keywords appearing simultaneously in the same document, and through hierarchical clustering, it exposes the affinity between keywords and the main theme [37]. Co-occurrence frequency and centrality are two key indicators, the former reflecting the intensity of attention to keywords and the latter indicating the importance of keywords in the co-occurrence network.

After conducting a co-occurrence analysis of keywords in 1434 English articles, 530 keyword nodes were obtained. In the keyword co-occurrence network, the size of each node represents the frequency of occurrence of the corresponding keyword. The larger the node, the higher the frequency of occurrence. As shown in Figure 2, keywords such as “physical activity”, “built environment”, “walking”, “health”, and “obesity” have larger nodes, indicating greater attention and making them the main research topics in English-language literature regarding the impact of street environment on physical activity. Hotspot analysis not only considers the frequency of keyword occurrence but also comprehensively evaluates keyword centrality. Based on the centrality of keywords shown in

Table 1. Top 20 hotspot words in English-language literature.

Serial Number	Keyword	Frequency of Occurrence	Centrality	Initial Appearance Year
1	physical activity	940	0.02	2002
2	built environment	604	0.02	2004
3	walking	389	0.03	2004
4	health	346	0.04	2002
5	obesity	204	0.03	2005
6	associations	179	0.01	2005
7	walkability	162	0	2009
8	adults	143	0.02	2005
9	travel	129	0.02	2005
10	design	120	0.02	2005
11	neighbourhood	116	0.02	2005
12	land use	108	0.04	2005
13	transportation	103	0.02	2005
14	urban form	103	0.02	2005
15	children	99	0.02	2005
16	older adults	98	0.03	2008
17	public health	94	0.07	2004
18	environment	93	0.03	2005
19	perceptions	91	0.03	2007
20	urban	89	0.02	2007

, it is evident that the keyword “public health” is the most significant (0.07), followed by “health” (0.04) and “land use” (0.04). This indicates their central roles in the co-occurrence network and widespread attention among international scholars conducting related research.

After conducting a co-occurrence analysis of 665 Chinese-language articles, a total of 372 keyword nodes were obtained. The co-occurrence frequency, ranked from high to low, includes “physical activity”, “built environment”, “landscape architecture”, “non-motorized transport”, “healthy city”, “elderly”, “public health”, urban design”, and so forth (as shown in Figure 3). Additionally, “physical activity” (0.37), “built environment” (0.16), “landscape architecture” (0.14), and “non-motorized transport” (0.15) are also significantly central keywords (as shown in

Table 2. Top 20 hotspot words in Chinese-language literature.

Serial Number	Keyword	Frequency of Occurrence	Centrality	Initial Appearance Year
1	physical activity	78	0.37	1994
2	built environment	67	0.16	2014
3	landscape architecture	35	0.14	2011
4	non-motorized transport	33	0.15	2008
5	healthy city	26	0.05	2013
6	elderly	24	0.03	2018
7	urban transportation	20	0.08	2010
8	public health	18	0.03	2015
9	urban design	16	0.07	2013
10	non-motorized transport system	14	0.06	2012
11	healthy	14	0.02	2010
12	children	12	0.02	2011
13	transportation planning	12	0.06	2008
14	public transportation	11	0.04	2005
15	traffic engineering	11	0.04	2010
16	green transportation	11	0.07	2010
17	risk factor	10	0.02	1997
18	urban renewal	9	0.06	2010
19	low-carbon transport	8	0.05	2011
20	influencing factor	8	0.01	2008

), indicating that these keywords are hotspots in Chinese-language literature regarding the impact of the street environment on physical activity.

3.2. Development Context

The study integrates the annual publication volume of English- and Chinese-language literature, distribution maps of keyword temporal trends, and keyword emphasis maps to comprehensively analyse the development context of research on the impact of street environments on physical activity. Academic output is one of the important indicators for assessing the development status of a research field, and it is significant for predicting development trends and dynamics [38]. From the statistical graphs of publication volume, English-language literature (as shown in Figure 4) shows an overall increasing trend in annual publications, entering a growth pattern in 2004, with fluctuating increases observed after 2016, and with a significant surge in 2019. Chinese-language literature (as shown in Figure 5) exhibits a gradual-to-rapid growth trend in annual publications, notably increasing in 2008, experiencing a slight decline in 2013, and followed by rapid growth leading to a peak in 2019. After 2019, the annual publication volume gradually decreased, contrasting with the trend observed in English-language literature publications. Through literature retrieval and review, it was found that the outbreak of the COVID-19 pandemic at the end of 2019 had an impact, with the Chinese government actively utilizing communities as basic governance units in epidemic prevention and control processes [39,40]. In the post-pandemic era, Chinese scholars have placed greater emphasis on healthy city construction, starting from community environments. This has led to a decrease in research publications focusing on urban street environments.

The timeline graph of keywords clearly demonstrates the evolution of keywords over different time spans and the interrelationships among them [41]. The burst detection graph of keywords with significantly increased frequencies over a short period reflects shifts and progress in research during different periods [42]. Therefore, by integrating the changes in the annual publication volume of English literature (as shown in Figure 4) and based on the timeline graph of keywords (as shown in Figure 6) and the burst detection graph of keywords (as shown in Figure 7), we summarize the evolutionary process of research on the impact of street environments on physical activity in developed countries. This process can be divided into two stages.

Phase 1 (2004–2016): Since the publication, in 2004, of the globally pioneering and influential “Streetscape Guidance: A Guide to Better London Streets” [43], scholars from various countries worldwide have gradually initiated research on street environments. In 2007, street space governance in New York focused on enhancing user interaction and practical exploration [44], adhering to the governance concept of “streets for all” [45], effectively meeting the needs of street users, and publishing the “Street Design Manual” in 2009. Until 2016, research in this phase was predominantly represented by American research teams, with a publication count of 397 papers, accounting for 54.6% of the total publications. The research focused on exploring decisive factors affecting human health during urban morphological changes and city expansion processes from a macro perspective. The burst keywords “urban form” and “urban sprawl” highlighted in Figure 7, along with the high-frequency keywords “land use” and “urban form” used since 2005, as indicated in Table 1, illustrate this focus. For example, Brownson evaluated the impact of urban built environments on physical activity from four aspects: land use, transportation, aesthetics, and safety [46]. Ewing and Cervero incorporated accessibility and distance factors into a 3D model (density, land use, and design), ultimately forming a widely used 5D model (density, land use, design, accessibility, and distance) for the quantitative analysis of street-based built environments [47]. These two highly cited papers in the field have profoundly impacted subsequent research, becoming the foundation for many scholars’ studies. Research methods often included interviews or questionnaires [48], systematic observation methods [46], and the use of high-resolution remote sensing imagery [49] for qualitative and quantitative analyses, dissecting the relationship between two-dimensional urban built environment characteristics and user activity behaviours.

Phase 2 (2016 to present): At the 9th Global Conference on Health Promotion in 2016, mayors from over 100 cities worldwide committed to prioritizing health in urban governance, taking measures to eliminate environmental pollution, improve urban infrastructure and public service systems, aiming to construct inclusive, safe, disaster-resistant, sustainable, and healthy cities. The Mayor’s Transport Strategy released in London in 2018 [50] and the Global Street Design Guide published in the same year in the United States [51] both proposed the concept of “healthy streets” and ten key indicators, summarizing in detail the research process on urban streets. The core principle of “prioritizing pedestrians in street design” has profoundly impacted the subsequent development of street design guidelines in many cities. Subsequently, scholars’ research perspectives shifted from a macroscopic view to focusing on the relationship between micro-level built environment elements of streets and physical activity from a human-centric perspective, expanding the research field centred on street space quality and accessibility. The burst keywords “quality” and “accessibility” from Figure 7 can attest to this shift. For instance, Yin objectively measured street-level urban design quality in Buffalo, New York, using 2D and 3D geographic information systems (GIS), and tested the correlation between micro-environmental characteristics of streets and observed pedestrian counts and walkability scores, aiding in a better understanding of the impact of street-level urban design features on walkability [52]. Larkin et al. employed open-source crowdsourced data, street view images, geographic information systems (GIS), and deep learning methods to identify architectural environmental factors associated with safety, vibrancy, and aesthetics perception in 56 cities, offering a new research method for measuring street space quality [53]. With advancements in measurement technologies (geographic information systems, global positioning systems, street view imagery, and machine learning), the proliferation of individual measurement instruments (eye-tracking devices, electroencephalograms, and virtual reality technologies), and the development of complex models, this phase of research began to emphasize the application of street-level big data, machine learning, and other new technologies to support quantitative indicator measurement, significantly improving research measurement efficiency and accuracy.

The keyword timeline graph (as shown in Figure 6) shows that research hotspot keywords were concentrated in the early part of the 21st century. The nodes of keywords such as “health”, “physical activity”, “walking”, and ”built environment” are more significant, transitioning in colour from purple to yellow, indicating sustained scholarly attention. These keywords are clustered under the “ecological model”, and the literature review reveals the profound impact of the “social-ecological theory model” on the study of street environments and physical activity relationships. The “social-ecological theory model” is a representative theory in studying human behaviour influenced by the environment, placing individuals at the centre of ecosystems and comprehensively considering influences on physical activity participation across individual, interpersonal, organizational, community, and policy levels, leading to a more comprehensive and detailed understanding of the factors influencing activity behaviours [54,55]. Since Corti [56] proposed the initial “social-ecological model” in 1998, scholars have subsequently conducted research and created enhancements based on this theoretical model: in 2003, Pikora [57] used the “social-ecological theory model” to establish a theoretical framework for the relationship between physical activity and the environment; in 2005, Zimring et al. [58] further refined the “social-ecological theory model” and studied its relationship with physical activity across four different spatial scales within cities, regions, neighbourhoods, and communities, deepening the study of environmental influences on physical activity and providing a more comprehensive theoretical framework. Subsequently, under the guidance of the “social-ecological theory model”, scholars have conducted a substantial amount of empirical research and validation and modification of the model [59,60]. As research content diversified, scholars shifted focus from macro-level material spatial elements, such as road network design [61], land use [62], and their relationship with physical activity, to specific micro-environmental elements, such as street furniture [63], streetscape greening [64], specific populations such as children [65] and older adults [66], and various types of street environment assessments, including health impact assessments [67] and walkability index evaluations [68]. Research has gradually integrated theories from other disciplines, such as Maslow’s hierarchy of needs theory [69], self-determination theory [70], attention restoration theory [71], and risk compensation theory [72], to further explore the mechanisms through which street environments influence physical activity.

The keyword time zone graph clearly displays the distribution of research hotspots during different time periods. Observing these hotspots can summarize the developmental trends of research content over time, which helps identify the distribution and evolution of research frontiers [73]. Combining the changes in the annual publication volume of Chinese-language literature mentioned above (as shown in Figure 5), along with the keyword time zone graph (as shown in Figure 8) and keyword burst detection graph (as shown in Figure 9), it is evident that research on the impact of street environments on physical activity in China can be broadly divided into three stages.

Phase 1 (1997–2009): China’s focus on healthy cities originated from public health concerns and corresponded with the developmental stage of China’s urban and rural socioeconomics. China’s focus on physical activity began with medical research on diseases such as obesity and diabetes, primarily involving interviews and surveys of patients within specific streets or neighbourhoods, exploring factors in urban environments that lead to chronic diseases, and identifying physical activity as a significant determinant of human health [74,75]. In this research direction, keywords such as disease prevalence, risk factors, and diabetes received prolonged attention, stemming from Chinese doctors and scientists proposing in the “Daqing Diabetes Prevention Study” that simple lifestyle interventions (diet and exercise management) could reduce the incidence of diabetes by 51% in specific high-risk populations [76]. The conclusion that “diabetes can be prevented” was quickly recognized worldwide, sparking global scholarly interest in diabetes prevention from 1997 onwards. Subsequently, discussions in the public health field shifted from passive disease treatment to proactive health interventions, with unprecedented attention paid to urban material space and social space as critical determinants of health impact [77].

Phase 2 (2010–2016): With the rapid urbanization in China, factors such as air pollution, traffic congestion, and road traffic injuries have steadily increased, posing significant health risks related to transportation. Consequently, the impact of transportation on health became a focal point for both the government and scholars during this phase. Beijing introduced the “Beijing Urban Pedestrian and Non-Motorized Vehicle Traffic System Design Guidelines (2010)” [78], and in the same year, the National Development and Reform Commission initiated pilot projects for national low-carbon provinces and cities in Beijing, emphasizing the establishment of an industrial system characterized by low carbon emissions and advocating for low-carbon green lifestyles and consumption patterns [79]. This phase highlights the urgent need to meet urban expansion needs, urban traffic decongestion demands, and urban climate regulation requirements by developing green travel environments. Encouraging urban residents to use walking or cycling as low-carbon transportation methods helps mitigate environmental issues and promotes human health. This has driven extensive research led by transportation engineering professionals to enhance urban public transportation and pedestrian systems. For example, Xiangmin et al. analysed the design of slow transportation systems in Canada and proposed planning strategies for China’s healthy city slow transportation systems [80]. Since then, the planning concepts of urban transportation practitioners and infrastructure development have entered a new era, shifting from a primary focus on main routes and long-distance, high-efficiency motorized travel to a concern for the local area’s healthy environment and quality, as well as the travel experience of different groups within the city.

Phase 3 (2016 to present): During this period, China’s urban development has transitioned from “expansion based on quantity” to a new era of “enhancement based on quality”, whereby street space construction has shifted from prioritizing speed to pursuing quality. Enhancing urban space quality to provide residents with a healthy and comfortable living environment has become a significant focus for scholars. Scholars in human settlements are increasingly exploring research on urban health planning and design strategies, emphasizing urban ecological construction, green infrastructure, urban green space accessibility, physical activity, urban open spaces, and forest healing [81]. In 2016, Shanghai issued the “Shanghai street design guideline”, the first guideline to systematically explore urban street design from the perspective of “complete streets” in China [82]. That same year, the State Council issued the “Healthy China 2030 Planning Outline”, emphasizing “popularizing healthy lifestyles, optimizing health services, and building a healthy environment”. Under the guidance of national policies, disciplines such as landscape architecture and urban planning have become the primary research forces. These focus on enhancing the quality of urban street spaces to attract people to engage in physical activities and promote public health development actively. They aim to explore the mechanisms by which street environment elements influence human behaviour, thus providing a theoretical basis for the practice of healthy street design. For instance, Chinese scholars Ying et al. [83] first proposed new approaches to analysing and planning the quality of street spaces in the context of new data environments; Leiqing et al. [84] and Yan et al. [85], respectively, introduced theories and methods for healing-oriented and healthy street design.

The development trajectory of research on the impact of China’s street environment on physical activity exhibits characteristics of different stages led by different disciplines. It began with a focus on chronic disease prevention and completed a transformation from passive disease treatment to active health behaviour interventions led by the medical field. Subsequently, with the rapid expansion and development of Chinese cities, urban environmental issues gained attention at the national strategic level. The academic community began to focus on street environment construction for human service, achieving a shift in thinking from “vehicle-centric” to “people-centric”, led by transportation engineering. Alongside increasing demands from residents for street space quality, the Chinese government actively promoted the construction of healthy urban environments. Currently, this effort is led by disciplines such as landscape architecture and urban planning in the field of human settlement environments, exploring how street environment elements influence human behaviour perception to provide scientific foundations for developing healthy city plans, improving urban environments, and promoting public health.

3.3. Theme Categories

Data extraction and analysis were conducted using the Bicomb2.0. Bicomb2.0 was utilized to extract 65 keywords with a frequency of 15 times or more from 1434 English-language articles. The keyword threshold was determined based on the Price’s Law formula, which was initially used to identify core authors within a research field and is now applied to define high-frequency keywords [86,87]. The calculation formula is as follows:

$$m=0.749\ast \sqrt{n_{\mathrm{m}\mathrm{a}\mathrm{x}}}=0.749\ast \sqrt{367}=14.348$$

(1)

where m represents the threshold boundary and n_max represents the maximum frequency count.

Based on this, 65 keywords with a frequency of ≥15 times were identified as high-frequency keywords (as shown in

Table 3. High-frequency keywords (≥15 occurrences) in English-literature.

Serial Number	Keyword	Frequency of Occurrence
1	built environment	367
2	physical activity	365
3	walkability	174
4	walking	166
5	obesity	75
6	environment	62
7	children	61
8	neighbourhood	59
9	urban design	53
10	GIS	47
11	exercise	46
12	older adults	43
13	public health	40
14	active transportation	34
15	urban form	31
16	active travel	31
17	neighbourhood	29
18	urban planning	29
19	health	29
20	cycling	29
21	google street view	28
22	active transport	28
23	mental health	28
24	geographic information systems	27
25	deep learning	26
26	transportation	26
27	safety	25
28	accelerometer	24
29	land use	24
30	China	23
31	green space	23
32	accessibility	23
33	pedestrians	23
34	urban health	22
35	active living	22
36	machine learning	21
37	pedestrian	21
38	youth	21
39	mobility	21
40	space syntax	21
41	crime	20
42	COVID-19	20
43	street greenery	20
44	parks	20
45	adolescents	19
46	open streets	19
47	adolescent	19
48	reliability	18
49	neighbourhoods	18
50	aging	18
51	gender	17
52	bicycling	17
53	transport	17
54	air pollution	17
55	recreation	17
56	physical environment	17
57	elderly	16
58	policy	16
59	social environment	16
60	urban environment	16
61	street connectivity	16
62	public space	16
63	neighbourhood environment	16
64	epidemiology	15
65	measurement	15

). Compared with the high-frequency keywords (as shown in Table 1) analysed by Citespace software above, it was found that, although the frequency of keywords was different due to the different file formats of Citespace and Bicomb analysis, the keyword segments were consistent, which verified the reliability of the analysis results.

Topic categories of the impact of the street environment on physical activity were analysed through cluster analysis using SPSS 27.0 software. Cluster analysis can aggregate topic groups representing certain research areas with solid interrelations among them [88], effectively compensating for the limitation of keyword co-occurrence networks in presenting connections between each pair of keywords.

Using Bicomb2.0, we constructed a dissimilarity matrix for 65 high-frequency keywords from English-language literature. This dissimilarity matrix was then imported into SPSS 27.0. Under the “Analysis” menu, we selected “Hierarchical Cluster Analysis” and chose “Within-groups linkage” as the clustering method, with measurement set to “Dice” under the “Binary” option. This process generated a dendrogram for the cluster analysis (as shown in Figure 10). Based on this analysis, the high-frequency keywords were grouped into nine clusters (as shown in

Table 4. Cluster analysis of keywords in English-language literature.

Cluster	Number of Keywords	Cluster Analysis
1	2, 3, 5, 4, 10, 6, 7, 11, 16, 8, 46, 18, 25, 29	built environment, physical activity, walking, walkability, urban design, obesity, environment, GIS, urban form, children, adolescents, neighbourhood, geographic information systems, accelerometer
2	9, 48, 54, 56, 21, 17, 19, 44	neighbourhood, adolescent, transport, recreation, cycling, active travel, urban planning, street greenery
3	12, 35, 14, 55, 27, 60, 51	exercise, urban health, public health, air pollution, transportation, social environment, aging
4	13, 45, 20, 28, 37	older adults, parks, health, safety, machine learning
5	49, 57, 15, 38, 23, 39, 61, 64	reliability, physical environment, active transportation, pedestrian, active transport, youth, urban environment, neighbourhood environment
6	47, 59, 36, 53, 58, 65	open streets, policy, active living, bicycling, elderly, epidemiology
7	22, 26, 33, 34	google street view, deep learning, accessibility, pedestrians
8	24, 32, 40, 31	mental health, green space, mobility, China
9	41, 66, 30, 63, 62, 42, 50, 43, 52	space syntax, measurement, land use, public space, street connectivity, crime, neighbourhoods, COVID-19, gender

). The thematic categories include the following:

(1)

Research on the walkability index of the urban built environment. This topic encompasses 14 high-frequency keywords such as “built environment”, “physical activity”, and “walking”. The focus of this theme revolves around the spatial quality of street-built environments, exploring the visual–spatial quality of built environments regarding urban vitality, public health, and environmental perception. Studies in this area investigate correlations such as walkability and urban vitality [89], walkability and public health [12], and the relationship between walkability and environmental perception [90]. Subjective variables in these studies often use social media, location-based services (LBS) data, or public ratings combined with machine learning methods for measurement. Objective variables frequently rely on OpenStreetMap (OSM) data, street view images, and semantic segmentation through deep learning to quantify visual elements of the built environment, thus constructing relevant indicators for spatial quality assessment. Notably, studies focusing on establishing direct or indirect relationships between urban greenery features and walkability using survey research, statistical analysis, or spatial analysis methods [64,91] are among the highly prioritized topics within this theme.

(2)

Exploring factors that promote active travel among the public. This topic consists of eight high-frequency keywords such as “neighbourhood”, “active travel”, and “urban planning”. The primary focus of this theme is to explain the relationship between public engagement in active travel (dependent variable) and social demographic and geographical factors (independent variables) through the construction of binary regression models. Scholars are particularly interested in adolescents, and the research area of significant interest is the concept of the 15 min neighbourhood [92,93].

(3)

Studying the impact of urban environmental hygiene on outdoor physical activity among the public. This topic includes seven high-frequency keywords such as “exercise”, “public health”, and “transportation”. Urban environmental hygiene primarily refers to natural and social environmental factors including air, water, soil, urban and rural planning, and living conditions. Research in this area focuses on issues such as climate change [94], sky view factor (SVF), an effective indicator describing urban radiation and the thermal environment [95], air quality [96], and other aspects of urban physical environments to analyse their impact on outdoor physical activity among the public.

(4)

Exploring street environment factors influencing physical activity among the elderly population. This topic primarily includes five high-frequency keywords such as “old adults”, “health”, and “safety”. Engaging in physical activity is an essential factor for promoting healthy ageing, and the elderly population is particularly susceptible to environmental influences due to declining physical function. Street traffic safety and personal safety related to crime are major factors affecting activity frequency among this demographic, drawing significant attention from scholars [31,97]. Research in this area primarily employs interviews, surveys, and behavioural investigations to identify the perceptions of older adults towards street spaces and the street-level elements impacting these perceptions. With rapid advancements in computer vision and deep learning technologies within the field of artificial intelligence, studies are now utilizing street view images and machine learning methods to identify street facilities and building facade elements affecting the perceived safety of elderly individuals. Examples include traffic signs [98], road surface conditions [99], vehicle traffic and parking situations [100], and buildings’ window-to-wall ratios [101], among other factors.

(5)

Conducting environmental audits focusing on the micro-scale landscapes of streets. This topic primarily includes eight high-frequency keywords such as “reliability”, “pedestrian”, and “youth”. After macro-scale built environment factors were shown to be correlated with physical activity, there has been continued contemplation among ordinary citizens and expert scholars on “what makes a good street landscape?” Many scholars have attempted to find scientific methods to assess pedestrian perspectives on street landscapes. Among these efforts, Microscale Audit of Pedestrian Streetscapes (MAPS) has emerged as an extensively validated observational field audit tool used in health research. MAPS is primarily employed to evaluate whether street environments support physical activity and to identify streetscape elements correlated with physical activity [102,103]. With the application of street view images, virtual audit tools like virtual assessment tools (Virtual-STEPS) [104] and Computer-Assisted Neighborhood Visual Assessment Systems (CANVAS) [105] are also being validated. Within this research theme, scholars focus significantly on adolescents, particularly assessing whether environmental elements along school-commuting routes attract adolescents to engage in physical activity voluntarily [106,107].

(6)

Exploring the impact of “Active Living by Design (ALbD)” on promoting community physical activity. This topic is primarily composed of six high-frequency keywords such as “open streets”, “active living”, and “bicycling”. ALbD is a national program initiated by the Robert Wood Johnson Foundation (RWJF) in 2002, which serves as a community grant program dedicated to exploring and improving environments that support active living. It aims to assist communities in creating environments that facilitate active living using the “social-ecological model” [108]. Research within this theme primarily utilizes experimental methods and has validated in multiple American cities that the ALbD community action model provides effective strategies (the 5Ps: preparation, promotion, planning, policy, and physical projects) to promote community physical activity and support community change [109,110,111,112].

(7)

Analysing landscape elements that influence pedestrian activity preferences using street view big data. This topic includes four high-frequency keywords: “Google Street View”, “deep learning”, “accessibility”, and “pedestrians”. The physical attributes and quality of urban environments can impact people’s perceptions of urban spaces. Given that panoramic images captured from a human perspective can comprehensively capture details within the environment, they have become important data sources for perception assessments. Studies of this nature typically involve offline surveys or crowdsourcing via the internet to pose brief evaluation questions about street view images, examining public perceptions of urban environmental quality [113]. Subsequently, computer vision techniques are employed to extract street view image features annotated with emotional biases, followed by the use of machine learning algorithms to analyse perceived attributes or types of image scenes [114]. This process aims to identify urban environments that better align with pedestrian preferences and behaviours.

(8)

Studying the impact of urban green spaces on people’s mental health. The high-frequency keywords include “mental health”, “green space”, “mobility”, and “China”. In recent years, the influence of urban green spaces on the physiological, psychological, and social health of the public, as well as the health benefits of green spaces, has received attention in international research on urban environmental health, urban planning, and urban sustainability fields. Regarding mental health, studies have shown that urban green spaces can alleviate stress among nearby residents [115] and reduce the risk of mental illness [116]. Such research is often based on theories related to the health benefits of nature, such as attention restoration theory (ART) [117], stress reduction theory (SRT) [118], and the biophilia hypothesis [119]. For data acquisition, methods such as emotion response scales, sustained attention response tests, stress perception scales, and self-assessment are used to measure psychological health outcomes [120]. Additionally, with advancements in experimental methods, researchers have begun using techniques like biofeedback devices, functional magnetic resonance imaging (fMRI), and electroencephalography (EEG) to assess physiological indicators such as blood pressure, blood flow, skin conductance, and salivary cortisol levels among participants [121,122]. Furthermore, many studies on this topic have focused on China as the research area [123,124,125].

(9)

Analysing the impact of urban built environment morphology on public behavioural activities. This topic includes nine high-frequency keywords such as “space syntax”, “land use”, and “crime”. Urban built environment morphology indicators encompass land use accessibility and the accessibility of street networks based on space syntax models. This theme focuses on studying the effects of green space accessibility [126], transportation infrastructure accessibility [127], street network accessibility [128], and land use characteristics [129] on public behaviour. This analysis can lead to insights into urban crime issues and the examination of spatial development equity differences within the same city or between different cities, enriching the theoretical framework of environmental justice and socioeconomic impact mechanisms.

Using Bicomb2.0, 42 keywords with a frequency of seven or higher were extracted from 665 Chinese-language articles. The threshold for these keywords was determined based on Price’s Law formula, which is calculated as follows:

$$m=0.749\ast \sqrt{n_{\mathrm{m}\mathrm{a}\mathrm{x}}}=0.749\ast \sqrt{81}=6.741$$

(2)

where m represents the threshold boundary and n_max represents the maximum frequency count.

Based on this, 42 words with a frequency of seven or more were determined to be high-frequency keywords in the Chinese-language literature, as shown in

Table 5. High-frequency keywords (frequency ≥ 7) in Chinese-language literature.

Serial Number	Keyword	Frequency of Occurrence
1	physical activities	81
2	built environment	68
3	landscape architecture	38
4	landscape architecture	38
5	elderly	27
6	healthy city	26
7	urban transportation	25
8	non-motorized transport system	21
9	urban design	20
10	public health	18
11	green transportation	17
12	health	15
13	urban renewal	14
14	planning strategy	13
15	urban plan	13
16	transportation planning	13
17	obesity	12
18	overweight	12
19	children	12
20	public transportation	12
21	sustainable development	11
22	traffic engineering	11
23	risk factor	11
24	outdoor physical activity	10
25	greenway	10
26	influencing factor	10
27	planning and design	9
28	accessibility	9
29	recreational physical activity	8
30	new urbanization	8
31	low carbon transportation	8
32	prevalence rate	8
33	community	8
34	city	8
35	rail	8
36	transit-oriented development (TOD)	7
37	big data	7
38	Beijing	7
39	slow travel space	7
40	green travel	7
41	chronic	7
42	ecology	7

.

Using Bicomb 2.0, a dissimilarity matrix was constructed for 42 high-frequency keywords extracted from Chinese-language literature. This dissimilarity matrix was then imported into SPSS 27.0 to generate the dendrogram for cluster analysis, as shown in Figure 11. Based on the dendrogram, the high-frequency keywords were clustered into five categories (as shown in

Table 6. Cluster analysis of keywords in Chinese-language literature.

Cluster	Number of Keywords	Cluster Analysis
1	18, 19, 24, 33, 27, 42	obesity, obesity, risk factor, prevalence rate, influencing factor, chronic
2	20, 25, 39, 34	children, outdoor physical activity, Beijing, community
3	11, 16, 6, 30, 2, 3, 13, 7, 29	public health, urban plan, elderly, recreational physical activity, physical activities, built environment, health, healthy city, accessibility
4	10, 38, 9, 26, 4	urban design, big data, non-motorized transport system, greenway, landscape architecture
5	15, 17, 21, 22, 12, 5, 8, 23, 36, 41, 37, 35, 28, 43, 14, 32, 31, 40	planning strategy, transportation planning, public transportation, sustainable development, green transportation, slow traffic, urban transportation, traffic engineering, rail, green trave, TOD, city, planning and design, ecology, urban renewal, low carbon transportation, new urbanization, slow travel space

). The thematic categories include the following:

(1)

Studies on the promotional effect of physical activity on chronic disease prevention. This topic is primarily composed of six high-frequency keywords, including “obesity”, “overweight”, and “risk factors”. The development of urbanization has led to changes in people’s lifestyle habits, resulting in an increasing prevalence of chronic diseases, particularly drawing attention from scholars to issues such as obesity and diabetes. Researchers mainly obtain population health indices and relevant influencing factors through questionnaire surveys and physical examinations, exploring urban natural environments (water, light, wind, etc.), built environments (aggregation of man-made constructions and modifications), and social demographic factors that trigger chronic diseases, thereby formulating intervention measures for chronic disease prevention and control [130]. The literature included in the analysis primarily focuses on selecting five healthy lifestyle factors: moderate alcohol consumption, non-smoking, maintaining a healthy weight (not overweight or obese), healthy diet, and regular physical activity [74,131]. Additionally, factors such as sleep [132], prolonged sitting [133], social interactions [134], and mental health [135] are also within the scope of scholars’ research.

(2)

Exploring community environments conducive to physical activity for children. This topic primarily composed of four high-frequency keywords: “children”, “outdoor physical activity”, “Beijing”, and “community”. Insufficient physical activity among children leads to physical deterioration, decreased independence, obesity, myopia, and other issues, drawing significant attention from scholars to children’s growth-related problems. Researchers such as Sisi et al. [136] have provided detailed reviews of the current research on urban children’s outdoor activity space allocation, exploring the impact of outdoor environments in different urban communities on children’s outdoor activities [137,138,139,140]. Children’s physical activity is closely linked to environmental spaces; actively enriching and natural environments can attract children to engage in a large amount of active play [141]. Existing studies often utilize methods such as tracking surveys [142], Global Positioning System (GPS), and accelerometers [143] to obtain children’s commuting routes and the surrounding sports venues and community facilities, using quantitative research methods to reveal the relationship between the two.

(3)

Studies on the impact of urban built environment on the health of elderly populations. This topic primarily composed of nine high-frequency keywords such as “public health”, “urban planning”, “elderly individuals”, and “leisure physical activities”. Against the backdrop of an ageing population, concerns arise about the health vulnerability of elderly groups. Various complex factors influence the health of elderly individuals. In addition to the threshold of activity capacity due to individual differences, there is also a strong correlation between urban-built environments and active physical activities among the elderly. In Chinese cities, particularly in Hong Kong, comprehensive research has been conducted on the influence of street environments on physical activities among elderly individuals [144,145]. Existing studies have found that the accessibility of infrastructure, street connectivity, and land use diversity are critical factors in stimulating elderly participation in walking and recreational activities [146,147], demonstrating that physical activities can serve as influential mediating variables in the impact of built environments on health. Linchuan et al. [148] demonstrated a non-linear relationship between street greening and walking behaviour among the elderly in Hong Kong. Furthermore, research has shown that objective-built environments affect the subjective perceptions of elderly individuals, thereby influencing their health behaviours [149], providing a clear insight into the mechanism through which objective-built environment elements impact the health of the elderly.

(4)

Using big data methods to analyse the impact of street environments on physical activity. This topic is primarily composed of five high-frequency keywords, including “urban design”, “big data”, and “non-motorized transport system”. The transformation and planning of urban streets towards a more humane and refined design pose higher demands for research and analysis on the human scale. The combination of street view images and machine learning enables the accurate deep processing of street view images, effectively identifying various street landscape elements such as sky, sidewalks, lanes, buildings, and greenery [150], improving the previous challenges of acquiring primary street data and efficiently utilizing street view images. Additionally, movement trajectories and check-in data from mobile applications provide substantial data support for scholars to understand public activity conditions [151], commonly used for studying residents’ exercise behaviour characteristics [152] and health level assessments [153]. Furthermore, the practice of utilizing big data to assist in planning and design is also beginning to emerge [154].

(5)

Exploring urban transportation systems that promote human health development. This topic primarily encompasses 18 high-frequency keywords such as “planning strategies”, “transportation planning”, and “public transportation”. Streets are essential components of urban transportation systems. Comprehensive urban transportation in China is currently undergoing a comprehensive transformation from focusing on speed to emphasizing development quality, from efficiency to resilience, from macro-level transportation and spatial structures to the integration of transit-oriented development (TOD) and station cities, and from a construction emphasis to comprehensive transformation through inventory renewal and retrofitting [155]. More and more scholars in transportation engineering and urban planning are paying attention to the impact of street environments on public health. Among the factors affecting public health in transportation systems, the expansion of road networks and the sharp increase in private vehicle ownership have led to significant issues of traffic noise and air pollution, which are important causes of chronic diseases [156]. The safety, connectivity, and walkability of streets not only promote physical activity but also have positive impacts on air quality and noise levels [157].

3.4. Research Trends

Multidimensional scaling (MDS) analysis was conducted using SPSS 27.0 software. Under the “Analysis” menu, the “Scaling” option was selected to access “Multidimensional Scaling”. The scaling model chose “Euclidean distance”. This produced quadrant diagrams representing the research hotspots on the impact of street environments on physical activity in international and Chinese contexts, as shown in Figure 12 and Figure 13.

The multidimensional scaling analysis presents a more intuitive graphical representation, displaying the relationships and similarities among keywords in a two-dimensional space. The closer the distance between keywords on the plane, the stronger the relationship between them [158]. The multidimensional scaling chart is divided into four quadrants counterclockwise. The origin represents the mean value; the horizontal axis (Dimension (1)) represents centrality and the external cohesion index, indicating the interaction strength of keywords with other keywords in different clusters; and the vertical axis (Dimension (2)) represents focus density and the internal cohesion index, indicating the internal association strength within the keywords composing the cluster, representing the degree of cluster development [159]. Keywords located in the first quadrant exhibit strong centrality and high density, occupying a central position in the research field. Keywords in the second quadrant, while not central topics, have high density, indicating relatively mature research. Keywords in the third quadrant have weaker density and insufficient centrality, typically indicating emerging or declining topics that are still immature in research. Keywords in the fourth quadrant exhibit strong centrality but lack internal maturity, requiring further in-depth research [160].

Figure 12. Multidimensional scaling plot of high-frequency keywords from English-language literature.

Figure 12. Multidimensional scaling plot of high-frequency keywords from English-language literature.

In the multidimensional scaling plot, clusters are represented by different colours and distributed across different quadrants based on their internal and external cohesion (referring to density and centrality, respectively). As shown in Figure 12, Cluster 1 in English-language literature, “Research on the walkability index of the urban built environment”, primarily occupies Quadrants I and II, indicating close internal connections within this research theme and extensive links with other themes, forming a particular research scale. Cluster 1 is closer to Clusters 2 and 3, suggesting relatively tight connections. Keywords like “neighbourhood”, “walking”, “environment”, and “obesity” are located in Quadrant I with high centrality, representing central concepts in this research theme. Cluster 2, “Exploring factors that promote active travel among the public”, predominantly occupies Quadrant I, signifying a highly central and dense topic indicating mature research with extensive connections to other themes. This cluster represents a core focus on the impact of street environments on physical activity research. Cluster 3, “Studying the impact of urban environmental hygiene on outdoor physical activity among the public”, shows dispersed keyword distribution across all four quadrants, indicating a loosely structured theme lacking systematic research. Keywords like “exercise”, “urban health”, “transportation”, and “elderly” are distributed in the first and fourth quadrants, indicating relatively high centrality. These keywords serve as central terms for this theme. Cluster 4, “Exploring street environment factors influencing physical activity among the elderly population”, mainly occupies Quadrants II and III with lower centrality. Research concerning elderly mobility safety is relatively mature, but studies using machine learning to identify street obstacles are less developed and require further exploration. Cluster 5, “Conducting environmental audits focusing on the micro-scale landscapes of streets”, primarily occupies Quadrants III and IV, indicating less depth in research. However, with the development and application of big data technologies, there is considerable potential for virtual auditing of street environments, warranting further investigation. Cluster 6, “Exploring the impact of “Active Living by Design (ALbD)” on promoting community physical activity”, mainly occupies Quadrants I and IV with higher centrality, suggesting ALbD has profound implications for subsequent research and public environment planning practices. Cluster 7, “Analysing landscape elements that influence pedestrian activity preferences using street view big data”, is positioned in Quadrant III with weaker centrality and density. Based on the initial appearance time of keywords, as shown in the timeline of keywords from the English-language literature in the knowledge graph (as shown in Figure 6), it is inferred that the research on this theme is in its initial phase, requiring scholars to conduct more systematic and in-depth studies. Cluster 8, “Studying the impact of urban green spaces on people’s mental health”, with its four keywords dispersed across Quadrants I, III, and IV, reflects limited systematic research on this topic. However, there is considerable interest among researchers, suggesting that there is room for further development. Cluster 9, “Analysing the impact of urban built environment morphology on public behavioural activities”, primarily occupies Quadrants III and IV with a less cohesive internal structure, indicating immature research. Based on the time of appearance of keywords (as shown in Figure 6), research in this category is relatively mature regarding urban crime issues. However, further deepening is needed, particularly concerning issues related to the fairness of urban resources.

Overall, conducting more in-depth and systematic research on urban street safety, environmental assessment, behavioural preferences, psychological healing, and issues of resource equity using extensive data analysis methods is the future trend of research in this field.

Figure 13. Multidimensional scaling plot of high-frequency keywords from Chinese-language literature.

Figure 13. Multidimensional scaling plot of high-frequency keywords from Chinese-language literature.

As shown in Figure 13, in Chinese-language literature, Cluster 1, “Studies on the promotional effect of physical activity on chronic disease prevention”, and Cluster 2, “Exploring community environments conducive to physical activity for children”, are located in the first quadrant. This indicates that these two topics have strong centrality and high density, suggesting mature research trends with extensive connections to other themes, becoming the core topics in the Chinese-language literature regarding the impact of street environments on physical activity. The keywords “chronic” and “influencing factor” in Cluster 1 and “community” and “Beijing” in Cluster 2 are closely linked. Cluster 3, “Studies on the impact of urban built environment on the health of elderly populations”, has its keywords distributed in the fourth quadrant, except for “urban planning”, indicating researchers’ interest but lacking internal maturity, requiring further study. The proximity of “public health” to “landscape architecture” in Cluster 4 and to “planning and design” in Cluster 5 indicates a close connection among them. Cluster 4, “Using big data methods to analyse the impact of street environments on physical activity”, has keywords scattered across all quadrants, indicating a lack of close connections and immature research. The keywords show that the theme revolves around research methods, specifically big data, with relatively close connections to Cluster 3 and Cluster 5. According to the time zone view of keywords (as shown in Figure 8), this cluster is inferred to be a newly emerging research topic that requires further deepening by scholars. Cluster 5, “Exploring urban transportation systems that promote human health development”, distributes keywords across the second, third, and fourth quadrants. Most keywords are located in the second quadrant, indicating close internal relevance of the theme and forming a particular scale of research, but with less connection to other themes. “Ecology” and “planning and design” are located in the fourth quadrant, serving as central terms for this theme. “Public transportation”, “sustainable development”, “urban renewal”, “green transportation”, and “traffic engineering” are in the third quadrant, representing future research trends for the theme.

Overall, utilizing extensive data methods to explore the impact of the urban built environment on the health of the elderly and employing extensive data methods to assist in constructing urban transportation systems to promote public health development are promising research directions for the future.

4. Comparative Analysis and Discussion

4.1. Development Context

Overall, the global research into and practice of healthy streets have transformed passive responses to survival challenges into proactive interventions for health benefits. A comparative analysis of research hotspots and development contexts in developed countries and China reveals differences in research perspectives, stages, theoretical foundations, and disciplinary backgrounds, while similarities are observed in research scales and methods (as shown in

Table 7. Comparative analysis of research between developed countries and China.

Research Comparison	Developed Countries	China
Research perspectives	Bottom-up	Top-down
Research stages	Post-assessment	Pre-planning
Theoretical foundation	Sociology: social–ecological theory model; psychology: Maslow’s hierarchy of needs, attention restoration theory, self-determination theory, risk compensation theory, and others.	Psychology: Maslow’s hierarchy of needs, attention restoration theory; environmental psychology theory.
Disciplinary background	human settlement environment, medicine, transportation engineering, social sciences, psychology, criminology, sports science, and others.	medicine, transportation engineering, and human settlement environment, and other.
Research scope	Macro-scale → human-centred scale	Macro-scale → human-centred scale
Research methods	Small data methods → big data methods	Small data methods → big data methods

).

Research perspectives: A comparison of high-frequency critical terms in research from developed countries (Table 1) and China (Table 2) reveals that, in addition to directly relevant keywords such as “physical activity”, “built environment”, and “public health”, high-frequency and high-centrality keywords in Chinese-language research also include “landscape architecture”, and “slow traffic”. Upon further examination of the relevant literature, it is found that the research perspective in China exhibits a “top-down” characteristic, with scholars predominantly guided by national policies. They analyse existing urban street design cases from a practical perspective or explore factors influencing physical activity through empirical research. Based on this, they provide optimization directions for design decisions, street design guidelines, transportation planning, and other policies, thereby guiding design implementation. Nevertheless, high-frequency keywords in international research also include “walking”, “obesity”, and “associations”. Combined with relevant literature, it is observed that research perspectives in developed countries demonstrate a “bottom-up” feature, with scholars focusing on developing targeted research questions from social life experiences. They primarily conduct causal mechanism studies on street space users, attempt to audit and evaluate already built street environments, and provide feedback on scientific issues, thereby influencing the formulation of urban planning and design policies.

Research stages: The differences in research perspectives indirectly reflect the different stages of research between China and developed countries. Research in China is mainly at the “pre-planning” stage, with the primary purpose being to provide the scientific basis for the planning guidelines and design decisions of healthy streets. Research in developed countries is mainly at the “post-assessment” stage, focusing on assessing and reviewing the healthy streets environment better to understand the mechanisms between street environments and physical activity.

Theoretical foundation: When reviewing the developmental context of research in developed countries and China, it is found that the theoretical foundation for environmental impacts on physical activity mainly originates from the Western academic realm’s “social-ecological theory model”. Internationally, this theoretical model has been utilized extensively for empirical research [59,60], demonstrating that multiple factors, including individual characteristics, material environmental factors, and social environmental factors, influence physical activity. However, Chinese scholars have applied this theoretical model less frequently, primarily remaining at the stage of literature review research [161,162,163] and lacking relevant empirical studies. Chinese scholars primarily rely on psychological theories such as Maslow’s hierarchy of needs [164], attention restoration theory [165], and Alfonso’s five-level system proposed in environmental psychology for walking experience—feasibility, accessibility, safety, comfort, and pleasure [166]—to conduct research on healthy streets environments. Due to the multitude of elements involved in the “social-ecological theory model” and the constantly changing factors affecting specific types of physical activity or under specific background conditions, there is often diversity in the influencing factors. Chinese scholars need to rely on the “social-ecological theory model” to explore the interactive effects and degrees of influence of objective material environments, social environments, and psychological and perceptual environments on behavioural activities within the Chinese context to clarify the direction of the theoretical model further.

Disciplinary background: Research in developed countries emphasizes interdisciplinary collaboration starting from disciplinary issues. For example, the most representative initiative is the “Active Living by Design” (ALbD) national program launched by the United States to address public health issues. Sponsored sites establish interdisciplinary community partnerships, including public health, urban planning, traffic engineering, architecture, pedagogy, and other disciplines, to collectively assess existing policies and environmental conditions, develop strategic plans, identify favourable resources, and collaborate with governments, space users, and others to achieve the sharing of academic research and social practice network resources [167]. In contrast, research in China is influenced by national strategic development, exhibiting a phased transition dominated by different disciplines with loose interdisciplinary cooperation. This is manifested by applying research methods from other disciplines while the research content still belongs to the respective disciplines, representing a borrowing relationship between disciplines [168]. For instance, sports science employs GIS research methods from urban planning to explore the relationship between the density, distance, accessibility, and physical activity of built environment facilities [169].

Research scope: By comparing the keyword timeline graphs of research in developed countries (Figure 6) and China (Figure 8), it is found that there is a trend towards transitioning from a macro-scale to a human-centred scale in the research scope. Specifically, there is a shift from focusing on the influence of aspects such as urban expansion, infrastructure, building density, land use mix, etc., on population physical activity [61,62,170] to paying attention to the impact of factors such as street design elements, street space scale, distance to public facilities, etc., on population physical activity [63,64,142].

Research methods: Both developed countries and China actively utilize multi-source and multi-scale big data, such as government open data, satellite remote sensing images, street view images, etc., combined with machine learning methods for large-scale data quantification analysis. However, research primarily focuses on the perspective of human visual perception, with fewer studies exploring the influence of olfactory and auditory landscape elements in street environments on physical activity. Moreover, existing empirical research mainly consists of static data analysis. It primarily explores the independent effects of single or multiple elements of the built environment on physical activity, neglecting the interactions between elements and temporal changes.

The reasons for the similarities and differences above are mainly reflected in the following three aspects: (1) Different urbanization development processes. Developed countries entered the late stage of urbanization by the end of the last century [171] and had earlier encountered urban public health challenges, leading to corresponding research and practices. On the other hand, China has been undergoing rapid urban expansion and development over the past 40 years, with a relatively late focus on constructing a healthy urban environment, resulting in a lag in research compared to developed countries. Furthermore, as China is in the transitional period of urban construction, many planning guidelines are needed, which require scientific theories and empirical support for their formulation. Additionally, the government attaches great importance to urban health planning and construction, proposing policies and initiatives to promote urban health development, including strengthening urban pollution control, increasing urban greenery levels, promoting nationwide fitness activities, and improving pedestrian transportation systems. These directly or indirectly contribute to the development of research on the impact of urban street environments on physical activity in China, thus presenting a “top-down” research characteristic. (2) Changes in urban planning perspectives. The implementation and management of urban design in various countries are shifting from the “two-dimensional plane” to the “three-dimensional space”, focusing on the human scale, which refers to the urban scale closely related to the human body that people can see, touch, and feel. This is a deepening and necessary supplement to the current scales such as grids, blocks, and plots [172]. (3) Development and application of new technologies. Big data methods such as street view images and machine learning, as well as the application of accelerometers and wearable devices, are less time-consuming and cheaper than traditional methods (questionnaires, behavioural observation), allowing for larger sample sizes and providing technical support for human-scale research.

4.2. Theme Categories

Research on the impact of street environments on physical activity in developed countries is categorized into nine themes: “Studies on the promotional effect of physical activity on chronic disease prevention”, “Exploring factors that promote active travel among the public”, “Studying the impact of urban environmental hygiene on outdoor physical activity among the public”, “Exploring street environment factors influencing physical activity among the elderly population”, “Conducting environmental audits focusing on the micro-scale landscapes of streets”, “Exploring the impact of “Active Living by Design (ALbD)” on promoting community physical activity”, “Analysing landscape elements that influence pedestrian activity preferences using street view big data”, “Studying the impact of urban green spaces on people’s mental health”, and “Analysing the impact of urban built environment morphology on public behavioural activities”. Research in China is categorized into five themes: “Studies on the promotional effect of physical activity on chronic disease prevention”, “Exploring community environments conducive to physical activity for children”, “Studies on the impact of urban built environment on the health of elderly populations”, “Using big data methods to analyse the impact of street environments on physical activity”, and “Exploring urban transportation systems that promote human health development”.

Research on the impact of urban street environments on physical activity fundamentally explores the causal relationship between the environment and individuals. Through the comparison of thematic categories, the differences in content between research in developed countries and China are analysed from two perspectives: “individual” (research subjects, activity categories) and “environment” (research areas, environmental factors) (as shown in Table 8). Regarding research subjects, studies in developed countries have a broader focus on the main population groups. In addition to the elderly [97], children [173], adolescents [174], and women [175], attention is also paid to special groups such as low-income groups and minorities [176,177]. Research in China primarily focuses more on the elderly [147], followed by children [142], and women have become the focus in recent studies [178], but there is less attention paid to adolescents, especially college students. However, statistics from the National Health Commission of China show that the prevalence of chronic diseases is becoming increasingly common among younger populations, with hypertension affecting 25% of residents aged 18 and above and abnormal blood lipids affecting 40% [179]. The health issues of adolescents are becoming increasingly prominent, requiring scholars to engage in targeted discussions about this population group. Regarding activity categories, research in developed countries categorizes physical activity in more detail. Currently, international physical activity studies are mainly categorized according to research purposes, with walking being one of the most common activities studied. It comprises the following four types of walking activities: work, transportation, leisure, household, and moderate-intensity physical activity [180]. In addition to walking, activities such as jogging [181], other sports [182], and cycling [183] are also studied, second only to walking. In Chinese-language research, physical activity is not subdivided in detail. However, there is a greater focus on walking activities, with research primarily focusing on the walkability [184], vitality [185], and spatial quality [172] of streets. Regarding research areas, developed countries typically divide study areas into urban areas [186], suburbs [187], and exurbs [188], considering factors such as the level of development, population density, housing type, and income. On the other hand, research in China mainly focuses on the urban core area as the research area [151], with few comparative studies combining different urban development regions. Regarding environmental factors, research in developed countries mainly focuses on the following four aspects: natural ecological environment (blue–green spaces, temperature, sunshine, etc.), the built environment (land use, sidewalk design, facility accessibility, etc.), the sociocultural environment (population density, crime density, etc.), and environmental psychological perception (sense of security, aesthetics, etc.). In contrast, Chinese-language research focuses more on the ecological environment, built environment, and environmental psychological perception elements, lacking investigation into sociocultural and environmental factors.

Table 8. Comparative analysis of research content between studies in developed countries and China.

Research Comparison		Developed Countries	China
Individual	Research subjects	Elderly, children, adolescents, women, low-income groups, and minorities	The elderly, children, and women
	Activity categories	Walking (for work, transportation, leisure, and household purposes), cycling, jogging, and other sports	Walking, cycling, and jogging
Environment	Research areas	Areas, suburbs, and exurbs, considering factors such as the level of development, population density, housing type, and income	Urban core area
	Environmental factors	The natural ecological environment (blue–green spaces, temperature, sunshine, etc.), the built environment (land use, sidewalk design, facility accessibility, etc.), the sociocultural environment (population density, crime density, etc.), and environmental psychological perception (sense of security, aesthetics, etc.)	The natural ecological environment, the built environment, and environmental psychological perception

Table 8. Comparative analysis of research content between studies in developed countries and China.

Research Comparison		Developed Countries	China
Individual	Research subjects	Elderly, children, adolescents, women, low-income groups, and minorities	The elderly, children, and women
	Activity categories	Walking (for work, transportation, leisure, and household purposes), cycling, jogging, and other sports	Walking, cycling, and jogging
Environment	Research areas	Areas, suburbs, and exurbs, considering factors such as the level of development, population density, housing type, and income	Urban core area
	Environmental factors	The natural ecological environment (blue–green spaces, temperature, sunshine, etc.), the built environment (land use, sidewalk design, facility accessibility, etc.), the sociocultural environment (population density, crime density, etc.), and environmental psychological perception (sense of security, aesthetics, etc.)	The natural ecological environment, the built environment, and environmental psychological perception

4.3. Research Trends

Based on the multidimensional scaling chart of research from developed countries and China (Figure 12 and Figure 13), as well as the comparative analysis of research hotspots, development context, and theme categories, the following three considerations and suggestions are proposed for the future research trends of China and other developing countries undergoing urbanization transition: (1) Macro level: National governments should further introduce planning guidelines with strong practical applicability and encourage interdisciplinary collaboration among urban planning, landscape architecture, medicine, transportation engineering, psychology, sociology, and other fields to discuss existing policies and environmental conditions. In addition, the government should encourage researchers, practitioners, and various stakeholders to jointly explore the best research schemes and pathways for healthy streets, effectively realizing the interaction between theory and practice. (2) Mesoscopic level: Research should focus on integrating local characteristics and shifting towards “localization” while fully drawing on research results and practical experiences from developed countries. Urban types and regions should be classified based on population size, geographical location, cultural functions, etc., and targeted assessments of the healthy street environment, should be conducted identifying the pain points and difficulties in street environment construction, and promoting the comprehensive implementation of public policies. (3) Micro level: Scholars should further study the causal mechanism between street environment elements and physical activity. Future research should, at the “individual” level, differentiate population attributes (children, elderly, disabled, special groups, etc.) and expand types of physical activity (running, cycling, leisure activities, etc.). At the “environmental” level, environmental factors from different sensory perspectives (olfactory comfort, auditory comfort, safety perception, etc.) and environmental characteristics of different street types (residential, transportation-oriented, commercial, etc.) should be examined for their impact on physical activity. Additionally, social environment factors such as social norms and community cohesion should be investigated. Integrated use of small data (surveys, interviews, participatory observations, etc.) and big data (street view images, urban points of interest, remote sensing images, etc.) methods should be emphasized to improve survey accuracy and effectiveness, promote the integration of temporal and spatial data with individual behaviour, and mainly focus on the long-term tracking and evaluation of influencing factors, examining the relationship between changes in built environment and physical activity for the same population.

5. Conclusions

The study selects China as representative of developing countries to conduct comparative research with developed countries. It uses relevant literature from the CNKI and WOS databases on the impact of street environments on physical activity as analysis samples. The research tools Citespace, Bicomb, and SPSS are utilized for quantitative and comparative analysis of research hotspots, development contexts, thematic categories, and research trends. The results indicate that, from a research perspective, China exhibits a “top-down” characteristic dependent on policy guidance, whereas developed countries display a “bottom-up” characteristic driven by scientific issues. In the research phase, studies in China are mainly in the “pre-planning” stage, while those in developed countries are primarily in the “post-assessment” stage. Regarding theoretical foundation, the theoretical basis for the impact of environment on physical activity mainly comes from the “social-ecological theoretical model” in Western academic circles, while Chinese scholars often introduce Maslow’s hierarchy of needs, attention restoration theory, etc. Regarding disciplinary background, research in developed countries focuses on interdisciplinary cooperation, starting from disciplinary issues, while research in China, influenced by national strategic development, shows a phased transition dominated by different disciplines with relatively loose disciplinary cooperation. On the research scale, both developed countries and Chinese-language studies show a trend of shifting from a macro-scale to a human-centred scale. In terms of research methods, scholars actively apply multi-source and multi-scale big data. Compared to Chinese-language studies, international research examines research subjects, activity types, regions, and environmental impact factors more meticulously and comprehensively in thematic categories. Based on this, the study summarizes three development directions for future research on the impact of street environments on physical activity in China: (1) Macro level: the government can encourage academic exchanges and practical activities on health street environment creation issues between disciplines and stakeholders to enhance the implementation of street planning guidelines; (2) Meso level: research and practice should integrate local characteristics and conduct targeted evaluations for different city types and regions to implement public policies fully; (3) Micro level: refine the influencing factors at the “human” and “environment” levels, conduct long-term tracking and evaluation, and further sort out the causal mechanisms between street environments and physical activity.

The study aids scholars in comprehensively understanding the current state and developmental directions of research on the impact of street environments on physical activity. From a comparative research perspective, it contrasts the similarities and differences between China and developed countries. From the perspective of street environments, it offers pathways and insights for addressing health issues in developing countries with backgrounds similar to China’s. It also assists international scholars in accurately understanding the status of research on the impact of street environments on physical activity within the Chinese context, thereby promoting global health welfare and equity. However, it is important to note that this article analyses the hotspots, contexts, themes, and trends of research on the impact of street environments on physical activity in China and developed countries solely from the perspective of “keywords”, lacking a comprehensive review of the global research network from the perspectives of “research institutions”, “authors”, and “publications”. Future research should integrate methods such as institutional network analysis, author collaboration network analysis, and co-citation analysis to present the core knowledge framework of the field of healthy street environments to readers.