Abstract
Previous studies have established that neighborhood environments influence mental health (MH), but the mediating role of physical activity in this relationship remains underexplored. Additionally, there is a lack of longitudinal research that investigates how neighborhood environments affect MH through physical activity pathways. We aimed to (1) analyze the cross-sectional relationships and longitudinal changes between neighborhood environments and MH, considering the mediating role of physical activity, and (2) investigate the urban-rural disparities in these associations. Data came from 15,470 adults in the 2010 and 2014 waves of the China Family Panel Studies. We applied multilevel structural equation modeling to examine cross-sectional and longitudinal relationships between neighborhood environments and MH via the pathway of physical activity. Cross-sectional results indicated that environmental aesthetics were positively associated with MH. The count of leisure facilities and environmental aesthetics were indirectly and positively associated with MH through participating in more physical activities. Meanwhile, there was no strong evidence on the association between longitudinal changes in neighborhood environments and longitudinal changes in MH. We also observed notable urban-rural variations in how neighborhood environments relate to MH. Environmental aesthetics were positively associated with MH in rural China, and physical activity did not mediate these associations. These findings suggested that improving access to sports and exercise facilities and enhancing environmental aesthetics and encouraging more physical activities could improve MH, especially in rural Northern China.
Supplementary Information
The online version contains supplementary material available at 10.1007/s11524-025-01025-3.
Keywords: Mental health, Neighborhood environments, Physical activity, Mediation pathways, Urban-rural disparities
Introduction
Mental health (MH) issues have been increasingly recognized as an important health risk due to their profound impact on overall well-being and quality of life [5]. It is necessary to take measures to promote MH. Multidimensional factors such as neighborhood environments [27, 44], physical activities [60], and sociodemographics [63] have been identified to be correlated with individuals’ MH. Comprehending the intricate connections among neighborhood environments, physical activity, and MH is essential for developing strategies that effectively enhance mental well-being.
Numerous cross-sectional and an increasing number of longitudinal studies have examined the associations between neighborhood environments and MH [33]. Recent cross-sectional and longitudinal systematic reviews have confirmed that neighborhood environments such as green space were linked to improved mental well-being [38, 52]. However, findings for other neighborhood environmental features remained inconsistent. For example, while some cross-sectional studies reported that greater access to retail outlets was related to better MH among older adults [64], a recent longitudinal study found a null association between the number of retail outlets and adults’ MH [32]. These inconsistencies may be attributable to differences in study design, geographic contexts, population characteristics, and neighborhood features [27].
Further, it is critical to examine the potential pathway of physical activity on the neighborhood environment–MH associations. Participating in physical activity could release endorphins, which could help lessen anxiety and depressive symptoms [60]. The social-ecological systems theory suggested that the neighborhood environment impacts MH directly by offering a pleasant environment and indirectly by promoting physical activity [36]. For instance, green spaces and environmental aesthetics provided attractive and accessible spaces that encouraged physical activity [24]. Neighborhood amenities like recreational and sports facilities provided spaces for exercise [65], promoting physical activity [18], and reducing stress [11]. Schools created environments that facilitated physical activities by offering opportunities for children and parents to engage in active routines and outdoor play [1], which helped establish healthier lifestyles. Retail outlets contributed to reducing stress by meeting daily needs efficiently and encouraging incidental physical activity, such as walking while shopping [48]. A growing number of studies have examined how physical activity mediates the neighborhood environments–MH associations [10, 30]. A few recent cross-sectional studies found that blue space proximity was correlated with MH through the pathway of participating in more physical activity in New South Wales, Australia [39] and in England [42]. Another cross-sectional study illustrated that a greener neighborhood was correlated with better MH in New Zealand, and physical activity mediated such association [47].
Despite growing evidence on neighborhood environment–MH associations accounting for the mediating role of physical activity, there were some limitations in the existing literature. First, previous studies mainly focused on blue/green space in developed countries without accounting for other neighborhood environment features like the availability of amenities. Neighborhood environments in China differed from those in developed countries in the availability of amenities and aesthetic quality [23]. Evidence incorporating multiple neighborhood environment features from large representative, multi-site studies in China that included all adult age groups remains rare. Second, limited empirical research explored rural-urban disparities on the neighborhood environments–MH associations and their potential pathway of physical activity. It is critical to examine rural-urban disparities due to the stark differences across urban, suburban, and rural areas in China [62]. The long-standing urban-rural dual structure and uneven pace of socioeconomic development have led to pronounced disparities in the availability of amenities and aesthetic quality across urban, suburban, and rural areas in China [34]. Urban areas feature a high concentration of public facilities and abundant leisure spaces. In contrast, rural areas often face shortages in basic infrastructure, limited availability of public facilities, and poor access to public transportation [34]. Suburban areas fall in between: they have lower availability of public facilities than urban areas but offer more facilities and better access to public transportation than rural villages [66]. In addition, mental health also varies among urban, suburban, and rural residents. For instance, one study found that rural residents in China reported poorer MH compared to their urban counterparts [13]. Rural residents may experience more social isolation due to limited transportation access or poorer social networks [15]. Physical activity patterns differed by urbanicity as well. Indeed, empirical evidence illustrated that urban residents tended to engage more in recreational or transport-related physical activity [43]. In contrast, rural residents did more physically demanding work such as farming but had fewer opportunities for recreational activity [31]. Meanwhile, suburban residents typically engaged in less transport-related activity than those in urban areas and less occupational activity than those in rural areas [68]. These differences may shape how neighborhood environments, physical activity, and MH are connected across urban, suburban, and rural areas [32]. Therefore, exploring rural-urban disparities extends insights into how broader structural and spatial contexts impact neighborhood environments–MH associations and their pathway of physical activity, underscoring the need for more equitable development strategies across rural and urban areas.
To address these gaps, we aimed to (1) investigate the potential pathway of physical activity on associations between neighborhood environments and MH, and (2) examine the rural-urban disparities in these relationships. This study contributes to the literature in two aspects. First, we provided a comprehensive understanding of neighborhood environment–mental health associations and their potential pathway of physical activity by accounting for multiple environmental features and using a large representative, multi-site dataset encompassing adults of all age groups in China. Second, we uncovered rural-urban disparities in the neighborhood environment-mental health associations considering the mediating role of physical activity.
Methods
Sample and Data Sources
The data were obtained from China Family Panel Studies (CFPS) [22], a longitudinal survey which was initially conducted in 2010, consisting of approximately 43,000 individuals. Biennial follow-up surveys have been conducted since. We focused on the surveys of 2010 and 2014, including 32,706 participants from 220 urban communities and 415 rural villages, as neighborhood environment information was only available in these 2 years [6]. Detailed sampling methodology of the CFPS was described in a prior study [32].
We adopted complete case analysis as in Besser et al. [4] to maintain comparability across survey waves and avoid hard-to-verify assumptions for imputing missing values [3, 57]. Although this approach reduced power and widened confidence intervals, coefficient estimates were consistent with multiple imputations [29], supporting the robustness of our findings. We retained 24,903 adult participants aged 18 or above at baseline for analysis. We further selected those with complete information on MH, neighborhood environment, physical activity, and sociodemographics in both 2010 and 2014. The analytical sample consisted of 15,470 participants (Supplement Fig. 2). Excluded participants were older, less frequently female, more likely to hold urban hukou, less often married or cohabiting, lived in urban areas, had higher household income, and lived in communities/villages with larger neighborhood population size (Supplement Table 1). Despite these differences, the sociodemographic characteristics of our analytical sample were largely similar to those in the 2010 Sixth National Population Census of China (data not shown) [49].
Variables
Outcome
We used the K6 scale survey questions to assess MH [20], which had adequate content validity [58] and sufficient reliability in China [26]. The detailed K6 scale survey question and the response options were displayed in Supplement Table 2. The questionnaire options ranged from 5 (“never”) to 1 (“almost every day”). We summed up the six response scores to assess the level of MH with scores ranging from 6 to 30 (higher values corresponded to better MH). In the cross-sectional analyses, the MH variable was continuous from the baseline survey in 2010. In longitudinal analyses, each participant’s MH score in 2014 was subtracted from the one in 2010 and categorized into a binary variable: stable/improved MH (≥ 0) and worsening MH (< 0). Specific details on the K6 scale and its validation can be found in a prior study [32].
Neighborhood Environments
Neighborhood environments include two components: neighborhood amenities and environmental aesthetics. Neighborhood amenities included four types of facilities: leisure facilities, shopping retailers, elementary schools, and healthcare facilities. Each type was measured as the count of the facilities in the community/village as reported by community/village officials. Environmental aesthetics was assessed by calculating the mean score of three survey statements: street cleanliness, architectural design, and ample space (higher scores correspond to a greater level of aesthetics). Details of neighborhood amenities and the three statements used to assess aesthetics were displayed in Supplement Table 3. In the cross-sectional analyses, the neighborhood environment variables were continuous from the baseline survey in 2010. In the longitudinal analyses, each neighborhood environment variable in 2014 was subtracted from the corresponding one in 2010 and categorized into a binary variable: stable/increased neighborhood environments (≥ 0) and decreased neighborhood environments (< 0).
Mediator
The mediating variable was physical activity, which came from the individual questionnaire in CFPS 2010. There was one survey question regarding the time used on physical activity: “During the past month, excluding vacations, how many hours per day on average did you engage in physical activities?”
Physical activity included walking, jogging, hiking, Tai Chi, dancing, aerobics, yoga, ball sports, water-based activities (e.g., swimming, diving), winter sports (e.g., skiing), and contact sports (e.g., wrestling, boxing) [16, 35]. In the cross-sectional analyses and the longitudinal analyses, the mediator was a continuous time-invariant variable from the CFPS 2010.
Covariates
In the cross-sectional analyses, the covariates included age, gender, Hukou type, BMI, level of education, marital status, household annual net income per capita (adjusted for inflation), ownership of cars, ownership of motorcycles, household size, count of children under 12 in the home, community type (urban/suburban/rural), daily hours of social engagement, and neighborhood population size, which all came from the baseline survey in 2010. In the longitudinal analyses, we included age, gender, and hours of social engagement per day at baseline, change in household size, change in count of children under 12 in the home, and change in the neighborhood population size, categorical household annual net income per capita (adjusted for inflation) in 2010 and 2014, categorical BMI in 2010 and 2014, categorical Hukou type (rural and urban) in 2010 and 2014, categorical marital status in 2010 and 2014, categorical education level in 2010 and 2014, categorical car ownership in 2010 and 2014, categorical motorcycle ownership in 2010 and 2014, and categorical community type in 2010 and 2014.
Statistical Analysis
We aimed to investigate how physical activity influences both cross-sectional relationships and longitudinal changes between neighborhood environments and mental health. Multi-level structural equation modeling (MSEM) was used to assess the cross-sectional and longitudinal associations [45]. Our conceptual framework constructed the pathway linking neighborhood environments to MH through physical activity (Fig. 1). Participants residing within the same neighborhood experienced the same neighborhood environments. Considering the nested data structure, we constructed a 2-1-1 multi-level mediation model (“2-1-1” represents the levels of the neighborhood environments, mediator, and outcome, respectively. The number “2” indicates level 2, and the number “1” indicates level 1). The neighborhood environments were between-level variables (level 2), while physical activity and MH were within-level variables (level 1). We conducted group-mean centering of the mediating variable at level 2 since neighborhood environments only varied at level 2. Additionally, we accounted for the influence of socioeconomic variables at level 1 mediating variable and the outcome variable.
Fig. 1.
Illustration of the pathway linking neighborhood environments and mental health
In the cross-sectional model, the variables for MH, neighborhood environments, and mediator were continuous in 2010. In the longitudinal model, we first calculated the change scores of the MH and neighborhood environments variables between 2010 and 2014, respectively. We further dichotomized each of the change scores in MH and neighborhood environments variables into two categories: < 0 indicated a decrease in MH or neighborhood environment and ≥ 0 indicated stable/increase in MH or neighborhood environment. The mediator was a continuous variable in 2010 for the longitudinal model. Besides, the covariates were from 2010 and 2014.
Further, we examined the urban-rural disparities in how neighborhood environments were related to MH and the mediating role of physical activity by stratifying the sample into three groups (urban/suburban/rural) based on community types reported in the CFPS questionnaires [32]. Urban areas referred to urbanized districts of prefecture- or county-level cities and township centers. Suburban areas referred to fringe areas at the edge of dense urban cores [67]. Rural areas referred to communities with mainly agricultural land use. Neighborhoods were categorized as urban, suburban, or rural by survey interviewers based on these definitions. We implemented the MSEM for each subgroup. We also conducted sensitivity analyses to investigate disparities between Northern and Southern rural regions [55], given their stark contrasts in weather, geographic context, and lifestyle.
We implemented the MSEM in Mplus 8.3 [40]. We chose maximum likelihood estimation to estimate the model parameters. In the cross-sectional models, we adopted commonly used indices to assess the model fit (see the footnotes of Table 3). We employed the maximum likelihood estimation to estimate the longitudinal MSEM models via logistic regression as the outcome variable was binary. The longitudinal MSEM also specified a random intercept, which had limited options for assessing SEM fit and lacked an external target value to evaluate the goodness of fit [9]. Therefore, we did not report model-fitting indices for longitudinal MSEM.
Table 3.
Adjusted cross-sectional and longitudinal association between neighborhood environments and mental health
| Sample size: 15,470 | Direct association | Indirect association via physical activity | Total association | |||
|---|---|---|---|---|---|---|
| Coefficient | 95% CI | Coefficient | 95% CI | Coefficient | 95% CI | |
| Panel A: cross-sectional results | ||||||
| Environmental aesthetics | 0.14** | (0.04, 0.24) | 0.004* | (0.00, 0.01) | 0.14** | (0.04, 0.24) |
| Count of leisure facilities | 0.04 | (−0.00, 0.08) | 0.004* | (0.00, 0.01) | 0.04* | (0.00, 0.08) |
| Count of elementary schools | − 0.06 | (− 0.16, 0.04) | − 0.001 | (− 0.00, 0.00) | − 0.06 | (− 0.16, 0.04) |
| Count of shopping retailers | 0.001 | (− 0.00, 0.00) | 0.000 | (− 0.00, 0.00) | 0.001 | (− 0.00, 0.00) |
| Count of healthcare facilities | − 0.02 | (− 0.08, 0.04) | 0.001 | (− 0.00, 0.00) | − 0.02 | (− 0.08, 0.04) |
| Panel B: longitudinal change results | ||||||
| Stable/improved environmental aesthetics | 0.01 | (− 0.15, 0.17) | − 0.003 | (− 0.01, 0.00) | 0.01 | (− 0.15, 0.17) |
| Stable/improved count of leisure facilities | − 0.02 | (− 0.16, 0.12) | − 0.001 | (− 0.02, 0.02) | − 0.03 | (− 0.17, 0.11) |
| Stable/improved count of elementary schools | − 0.03 | (− 0.19, 0.13) | 0.000 | (− 0.00, 0.00) | − 0.03 | (− 0.19, 0.13) |
| Stable/improved count of shopping retailers | 0.02 | (− 0.10, 0.14) | 0.000 | (− 0.00, 0.00) | 0.02 | (− 0.10, 0.14) |
| Stable/improved count of healthcare facilities | 0.09 | (− 0.07, 0.25) | 0.000 | (− 0.00, 0.00) | 0.09 | (− 0.07, 0.25) |
* < 0.05; ** < 0.01; *** < 0.001.
All Model adjusted age, gender, BMI, marital status, education level, Hukou type, household annual net income per capita, family size, car ownership, motorcycle ownership, household child count, community type, hours of social engagement per day, and neighborhood population size.
Model fit indices for cross-sectional analysis were: df = 1, Chi-square = 1.76, Chi-square/degree of freedom = 1.76 (≤3 indicates a good fit; values up to 5 suggest an acceptable fit), CFI = 1.000, TLI = 0.979 (CFI and TLI closer to 1 indicate a better fit), SRMR Value for Between = 0.001, RMSEA = 0·007 (< 0.05 indicate a better fit), SRMR Value for Within = 0.002 (< 0.05 indicate a better fit)
Results
Descriptive Results
Table 1 displays descriptive statistics of the sample. The mean participants’ age at baseline was approximately 46 years (standardized deviation [SD] = 14), with a balanced gender distribution (50.14% male). Overweight/obesity prevalence increased from 2010 to 2014 (30 to 35%). More than 70% of the participants had rural hukou in 2010 (73%), with a slight decline in 2014 (70%). The mean value of household annual net income per capita increased from 8,808 Chinese Yuan in 2010 to 12,798 Chinese Yuan in 2014. Around 87% of participants were married or cohabitated in 2010 and 2014. About 94% of participants had a high school degree or below in 2010. Car ownership increased from 13% in 2010 to 17% in 2014. Neighborhood population size increased slightly during 2010–2014.
Table 1.
Descriptive statistics of sample characteristics (N = 15,470)
| Sociodemographic characteristics | CFPS data | |
|---|---|---|
| 2010 | 2014 | |
| Age at baseline: mean (SD) | 46.51 (14.08) | 46.51 (14.08) |
| ≥ 18 and < 60: % | 80.92 | 80.92 |
| ≥ 60: % | 19.08 | 19.08 |
| Gender: % | ||
| Male | 50.14 | 50.14 |
| Female | 49.86 | 49.86 |
| Hukou type: % | ||
| Rural | 72.82 | 70.47 |
| Urban | 27.18 | 29.53 |
| BMI a: % | ||
| Normal | 61.62 | 57.27 |
| Underweight | 8.14 | 7.60 |
| Overweight | 24.74 | 27.84 |
| Obese | 5.49 | 7.30 |
| Education level: % | ||
| High school or below | 94.19 | 93.82 |
| Associate degree | 3.78 | 4.01 |
| Bachelor’s degree or above | 2.04 | 2.17 |
| Household annual net income per capita (Chinese yuan): mean (SD) | ||
| Q1 | 1791.44 (806.90) | 1997.77 (1251.11) |
| Q2 | 4293.77 (750.58) | 6731.82 (1251.45) |
| Q3 | 7687.99 (1301.59) | 12,595.63 (1974.47) |
| Q4 | 21,454.09 (21,456.41) | 29,445.34 (29,908.86) |
| Car ownership: % | ||
| No car | 86.83 | 83.06 |
| At least one car | 13.17 | 16.94 |
| Motorcycle ownership: % | ||
| No motorcycle | 50.88 | 50.35 |
| At least one motorcycle | 49.12 | 49.65 |
| Marital status: % | ||
| Cohabitation and married | 87.48 | 87.58 |
| Never married | 6.84 | 7.52 |
| Divorced or widowed | 5.68 | 4.90 |
| Community type: % | ||
| Urban | 27.72 | 26.30 |
| Suburban | 6.94 | 4.51 |
| Rural | 65.34 | 69.19 |
| Household size (person): mean (SD) | 4.20 (1.78) | 4.17 (1.93) |
| Hours of social engagement/day | 1.66 (2.09) | - |
| Household child count (person): mean (SD) | 0.59 (0.82) | 0.41 (0.70) |
| Neighborhood population size (person): Mean (SD) | 3779 (4159.20) | 4122 (4641.15) |
a. BMI were categorized into four categories: underweight (<18.5), normal (18.5–24), overweight (24–28), and obese (≥28) (China Ministry of Health Disease Control Division, 2003).
Table 2 displays the descriptive statistics of MH and neighborhood environments (2010 and 2014), and mediating variables (2010). MH experienced a slight decrease with the mean value of 26.97 in 2010 to 26.69 in 2014. Environmental aesthetics improved from 2010 to 2014 (mean: 4.4 to 4.7). The availability of leisure facilities showed a slight rise during the study period (mean: 1.8 to 2.1). The count of shopping retailers experienced a slight decline (mean: 25 to 24). The count of elementary schools remained relatively stable with a minor decrease (Mean: 1.60 to 1.55). The average number of healthcare facilities increased notably with a mean value of 2.3 in 2010 and 3.3 in 2014.
Table 2.
Distribution of mental health, neighborhood environments and mediator variable at baseline and follow-up (N = 15,470)
| CFPS data | |||||
|---|---|---|---|---|---|
| 2010 | 2014 | ||||
| Outcome | Mean (SD) | Mean (SD) | |||
| Mental health (Range: 6–30) | 26.97 (3.82) | 26.69 (4.06) | |||
| Change in mental health (continuous) during 2010–2014 | Range (min–max) | Median | Mean | 25th-75th percentile | SD |
| − 24 to 24 | 0.00 | − 0.28 | − 2.00 to 2.00 | 4.64 | |
| Neighborhood environments | Mean (SD) | Mean (SD) | |||
| Environmental aesthetics | 4.41 (1.19) | 4.70 (1.19) | |||
| Count of leisure facilities | 1.78 (2.59) | 2.06 (2.73) | |||
| Count of elementary schools | 1.60 (1.64) | 1.55 (1.69) | |||
| Count of shopping retailers | 24.88 (70.78) | 23.76 (99.14) | |||
| Count of healthcare facilities | 2.28 (2.49) | 3.32 (3.54) | |||
| Change in neighborhood environments (continuous) during 2010–2014 | Range (min–max) | Median | Mean | 25th–75th percentile | SD |
| Change in environmental aesthetics | − 4.66 to 5.00 | 0.33 | 0.28 | − 1.00 to 1.33 | 1.63 |
| Change in count of leisure facilities | − 19 to 15 | 0.00 | 0.27 | − 1.00 to 1.00 | 2.74 |
| Change in count of elementary schools | − 26 to 20 | 0.00 | − 0.05 | − 1.00 to 0.00 | 1.67 |
| Change in count of shopping retailers | − 670 to 1069 | 0.00 | − 1.31 | − 4.00 to 2.00 | 91.07 |
| Change in count of healthcare facilities | − 24 to 22 | 1.00 | 1.04 | 0.00 to 2.00 | 3.00 |
| Mediator Variable | Mean (SD) | - | |||
| Physical activity (h/day) | 0.52 (1.22) | - | |||
Cross-sectional Relationships and Longitudinal Changes Between Neighborhood Environments and Mental Health
Table 3 presents the MSEM results of neighborhood environments on adults’ MH in both cross-sectional and longitudinal analyses. The model fitting indices (Table 3) indicated a satisfactory fit. All models were controlled for sociodemographic variables. Supplement Table 4 presents the regression results of the full multi-SEM model with covariate adjustments.
In cross-sectional results (Table 3 panel A and Fig. 2), environmental aesthetics were positively associated with MH. Indirect associations were observed for the count of leisure facilities and environmental aesthetics through physical activity. Environmental aesthetics and the count of leisure facilities showed positive total associations with MH. Meanwhile, the count of elementary schools and healthcare facilities showed no significant direct, indirect, or total associations with MH.
Fig. 2.
Standardized direct associations of neighborhood environments on mental health from cross-sectional analyses
The longitudinal analyses (Table 3 panel B) revealed insufficient evidence to support a correlation between changes in neighborhood environments and changes in MH. For instance, living in areas with stable or increased levels of environmental aesthetics was not correlated with stable/improved MH.
Urban-Rural Disparities in the Relationships Between Neighborhood Environment and Mental Health
Table 4 presents variations in cross-sectional relationships between neighborhood environments and MH by urbanicity. Interestingly, these associations were exclusively evident in rural areas. In rural areas, environmental aesthetics were positively associated with MH, with a similar total association. Meanwhile, the count of elementary schools, the count of shopping retailers, the count of healthcare facilities, and the count of leisure facilities showed no significant associations with MH regardless of the community type of area. We further analyzed regional heterogeneity in rural areas between the Northern and Southern regions (Supplement Table 5). The cross-sectional results showed that in Northern rural areas, environmental aesthetics were positively associated with MH, with consistent total associations. The count of elementary schools was negatively associated with MH in Southern rural areas.
Table 4.
Urban-rural variations in the cross-sectional relationships between neighborhood environments and mental health
| Sample size: 15,470 | Direct association | Indirect association via physical activity | Total association | |||
|---|---|---|---|---|---|---|
| Coefficient | 95% CI | Coefficient | 95% CI | Coefficient | 95% CI | |
| Rural | ||||||
| Environmental aesthetics | 0.15* | (0.01, 0.29) | 0.001 | (−0.001, 0.003) | 0.15* | (0.01, 0.29) |
| Count of leisure facilities | 0.08 | (− 0.02, 0.18) | 0.001 | (− 0.002, 0.004) | 0.08 | (− 0.02, 0.18) |
| Count of elementary schools | − 0.14 | (− 0.32, 0.04) | − 0.001 | (− 0.002, 0.000) | − 0.14 | (− 0.32, 0.04) |
| Count of shopping retailers | − 0.001 | (− 0.01, 0.00) | 0.000 | (− 0.001, 0.001) | − 0.001 | (− 0.01, 0.01) |
| Count of healthcare facilities | − 0.05 | (− 0.17, 0.07) | 0.000 | (− 0.001, 0.001) | − 0.05 | (− 0.17, 0.07) |
| Suburban | ||||||
| Environmental aesthetics | 0.08 | (− 0.27, 0.43) | − 0.02 | (− 0.05, 0.01) | 0.07 | (− 0.27, 0.41) |
| Count of leisure facilities | 0.04 | (− 0.14, 0.22) | 0.02 | (− 0.02, 0.06) | 0.05 | (− 0.11, 0.21) |
| Count of elementary schools | − 0.14 | (− 0.39, 0.11) | − 0.03 | (− 0.07, 0.01) | − 0.17 | (− 0.43, 0.09) |
| Count of shopping retailers | − 0.01 | (− 0.03, 0.01) | 0.001 | (− 0.001, 0.003) | − 0.01 | (− 0.03, 0.01) |
| Count of healthcare facilities | 0.11 | (− 0.09, 0.31) | 0.02 | (− 0.01, 0.05) | 0.12 | (− 0.08, 0.32) |
| Urban | ||||||
| Environmental aesthetics | (− 0.08, 0.28) | 0.001 | (− 0.003, 0.005) | 0.10 | (0.08, 0.28) | |
| Count of leisure facilities | 0.01 | (− 0.03, 0.05) | 0.002 | (− 0.000, 0.004) | 0.02 | (− 0.02, 0.06) |
| Count of elementary schools | − 0.02 | (− 0.10, 0.06) | − 0.001 | (− 0.003, 0.001) | − 0.02 | (− 0.10, 0.06) |
| Count of shopping retailers | 0.001 | (− 0.00, 0.00) | 0.000 | (− 0.001, 0.001) | 0.001 | (0.00, 0.002) |
| Count of healthcare facilities | − 0.01 | (− 0.07, 0.05) | − 0.001 | (− 0.003, 0.001) | − 0.01 | (− 0.07, 0.05) |
*< 0.05; **< 0.01; ***< 0.001
All Model adjusted age, gender, BMI, marital status, education level, Hukou type, household annual net income per capita, household size, car ownership, motorcycle ownership, household child count, hours of social engagement per day, and neighborhood population size
The longitudinal analyses (Supplement Table 6) revealed insufficient evidence to support a correlation between changes in neighborhood environments and changes in MH, regardless of the community type of area.
Discussion
This study aimed to investigate cross-sectional relationships and longitudinal changes between neighborhood environments and MH, accounting for the pathway of physical activity. Cross-sectional analyses showed that environmental aesthetics had positive direct associations with adults’ MH. Furthermore, environmental aesthetics and the count of leisure facilities were indirectly correlated with MH via physical activity. Longitudinal analyses showed that living in areas with stable or increased neighborhood environments was not correlated with stable or improved MH. Physical activity did not mediate the longitudinal associations either. Additionally, urban-rural differences were evident in the cross-sectional relationships linking neighborhood environments to MH.
Firstly, we observed that higher levels of environmental aesthetics were positively associated with better MH in the cross-sectional results in China. Physical activity reinforced these positive associations. This aligns with earlier cross-sectional research showing that environmental aesthetics and perceived greenness promoted MH through having more recreational walking [51]. An appealing living environment with clean roads, a tidy architectural layout, and spaciousness could provide ample opportunities to encourage more outdoor physical activities and exercises [39], reducing perceived barriers to participating in sports and exercises, and fostering a sense of safety and comfort, ultimately contributing to improved MH [12]. Moreover, participating in more physical activity strengthens the positive relationships between MH and environmental aesthetics. These findings suggest that policymakers could prioritize strategies to improve environmental aesthetics to create visually appealing environments with clean streets and pleasing architectural designs.
Secondly, the cross-sectional results demonstrated that the count of leisure facilities was positively associated with better MH in China. Leisure facilities could promote MH by providing means for emotional regulation, enhancing social interaction, fostering self-confidence and self-efficacy, alleviating stress, and encouraging healthy lifestyle habits [63]. These findings support the hypothesis that more leisure facilities are linked to better MH [25]. Moreover, we found that the count of leisure facilities was indirectly correlated with MH through participating in more physical activities in cross-sectional analyses. Higher availability of leisure facilities could provide more opportunities for adults to engage in physical activity, which in turn contributes to improved MH. First, empirical evidence has illustrated that higher availability of physical activity facilities was correlated with higher frequencies of adults’ sports and exercise behaviors [25]. Second, the mechanisms by which sports and exercise behavior promote MH could be explained from both physiological and psychological perspectives. From a physiological perspective, exercise reduces stress hormones such as cortisol and adrenaline while increasing levels of endorphins, leading to feelings of happiness, optimism, and relaxation [41]. Additionally, exercise increases brain-derived neurotrophic factors (BDNF), which may alleviate symptoms of anxiety [17]. Psychologically, engaging in exercise can build self-efficacy, enhancing feelings of control over stressful situations [46]. Furthermore, exercise serves as a distraction from stressors, effectively reducing anxiety [37]. Overall, our results suggest that policymakers could develop intervention strategies to enhance access to leisure facilities and encourage more physical activities, which are conducive to promoting adults’ MH.
Further, we revealed urban-rural variations in how neighborhood environments relate to MH. Our results illustrated that environmental aesthetics significantly improved MH in rural China. This finding was consistent with previous findings indicating the relationship between environmental aesthetics and MH was more pronounced in rural China [32]. In China, many villages had fewer services and formal leisure spaces, and residents often lived in less appealing environments. This scarcity may have strengthened the association between environmental aesthetics and MH in rural China and made aesthetics an important resource for stress relief [8]. Meanwhile, physical activity did not mediate the association. One potential explanation was that we focused on the overall physical activity level rather than specific types due to data unavailability, which may attenuate the ability to detect the mediating role [19]. We further noted regional disparities, as the connection between environmental aesthetics and MH was evident exclusively in Northern rural China. This may have been due to differences in cultural backgrounds and community activity patterns between Northern and Southern rural areas [55]. Due to a less clean environment with more severe air pollution in Northern rural areas, residents may have a higher demand for attractive neighborhood environments, which could directly enhance MH [53]. Physical activity did not mediate such an association in Northern rural China. The lower temperature in Northern rural areas may limit residents’ outdoor physical activity, reducing its potential mediating association [59]. Overall, our results suggest that policymakers should enhance environmental aesthetics in rural areas by improving cleanliness, architectural layout, and spaciousness to support adults’ MH, especially in Northern rural regions.
Besides, longitudinal analyses revealed that the associations did not persist when examining changes in neighborhood environments and changes in MH over time in China. This result resonates with a prior study illustrating that restaurant density was correlated with lower dietary quality only in cross-sectional analyses, but not in longitudinal analyses [2]. It is likely due to the fact that variations in neighborhood environments and MH were minimal over a short period (4 years). Additionally, physical activity failed to mediate the observed relationships. This could be due to minimal changes in these variables during the study period [44]. Further, physical activity was measured at baseline (2010) only due to data unavailability in 2014. This temporal mismatch may attenuate indirect associations between neighborhood environments and MH in the longitudinal analyses. Nevertheless, our findings highlight that future research with an extended observational period is essential to investigate the potential mediating pathways and the longitudinal connections between neighborhood environments and MH. Future research could also focus on those who experience dramatic changes in living environments due to relocation (i.e., rural immigrants living in urban areas) and those living in rapidly evolving environments.
Strengths and Limitations
This study presents several strengths. Firstly, we accounted for the pathway of physical activities in the neighborhood environments—MH relationships, which could enhance our understanding of how neighborhood environments influence health. Second, our study conducted both cross-sectional and longitudinal analyses from a multi-site longitudinal survey with a substantial sample size in China, allowing us to examine both cross-sectional relationships and temporal changes between neighborhood environments and MH. Third, we examined urbanicity disparity in the cross-sectional analyses, which provides a more nuanced understanding of how neighborhood environments influence MH across different geographic contexts. Further, we used multilevel structural equation modeling, which was advantageous in examining multiple pathways on data with hierarchical structure [14]. The multilevel mediation model was able to consider the hierarchical structure of the data, thus providing more robust connections linking neighborhood environments to MH [45].
However, this study also has some limitations. One limitation is the relatively short-time period (4 years) in our longitudinal analyses, spanning from 2010 to 2014. This limited timeframe constrains our ability to capture long-term trends and changes in neighborhood environments and their longitudinal relationships with changes in MH. Future research could extend the observation period to obtain a more comprehensive understanding of longitudinal relationships over time. Second, we did not include changes in the mediating variable during 2010–2014, and physical activity was measured as a composite indicator without classification by intensity level (e.g., low, moderate, or high) [50] or type (e.g., functional or leisure) [21, 42] due to data unavailability. Future work could incorporate changes in physical activity, distinguishing both intensity and type, to improve insights into the longitudinal relationships linking neighborhood environments to MH via diverse physical activity forms. Third, the spatial scale effects of the built environment on MH were not examined due to the lack of geographic data. Future studies could investigate how this relationship varies across spatial scales [28]. Moreover, the influence of residential self-selection on neighborhood environments and MH remains unexplored because individual residential preference data were unavailable. Incorporating such preferences in future research could address potential self-selection bias. Besides, we did not establish causal relationships between neighborhood environments and MH. Future studies could adopt causal mediation approaches to address intermediate confounding and clarify causal pathways [7]. In addition, we used complete case analysis to retain participants who had complete information across all the variables. Future research could retain larger samples by adopting multiple imputations to handle missing data.
Conclusions
This study offers meaningful perspectives on how neighborhood environments are linked to MH in China. Our cross-sectional analyses revealed significant positive total associations of the count of leisure facilities and environmental aesthetics with MH. The cross-sectional results suggested that enhancing environmental aesthetics and improving access to sports and exercise facilities could potentially improve adults’ MH. Additionally, physical activities also mediated the associations linking neighborhood environments to MH, suggesting that participating in more physical activities is a promising strategy to promote MH. Besides, urban-rural disparity was observed in the cross-sectional links between neighborhood environments and MH. Environmental aesthetics were positively associated with MH in rural areas. These findings underscore the necessity of incorporating urban-rural differences when assessing the influence of neighborhood environments on MH. Meanwhile, changes in neighborhood environments and changes in MH during 4 years of follow-up data were minimal, constraining our ability to detect longitudinal associations with changes in MH. Future work could focus on those experiencing dramatic changes in living conditions due to relocation as well as those living in rapidly changing environments with an extended observational period.
Supplementary Information
Below is the link to the electronic supplementary material.
(PDF 259 KB)
Acknowledgements
This work was supported by the National Natural Science Foundation of China [grant numbers 42401235, 42271433], the Outstanding Young Talent of Wuhan Talent Program (202417100095), and the Early Career Research Grant by the School of Public Administration at China University of Geosciences (CUGGG-2308), and Fundamental Research Funds for the Central Universities (413000159, 2042024kf1024, 2042023kfyq04).
Author Contributions
Yanxiao Liu: conceptualization, data curation, methodology, formal analysis, visualization, writing—original draft. Jingjing Li: conceptualization, methodology, investigation, supervision, writing—review and editing. Peng Jia: writing—review and editing.
Footnotes
Highlights
• Cross-sectional relationships and longitudinal changes between neighborhood environments and mental health
• Exploring how physical activity mediates the connection between neighborhood environments and mental health
• Environmental aesthetics was positively associated with mental health in the cross-sectional results
• Leisure facilities and environmental aesthetics were positively associated with mental health through physical activities in the cross-sectional results
• Environmental aesthetics was positively associated with mental health in rural China, with no indirect association via physical activity
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Alme H, Reime MA. Nature kindergartens: a space for children’s participation. J Outdoor Environ Educ. 2021;24(2):113–31.
- 2.Auchincloss AH, Li J, Moore KA, Franco M, Mujahid MS, Moore LV. Are neighbourhood restaurants related to frequency of restaurant meals and dietary quality? Prevalence and changes over time in the Multi-Ethnic Study of Atherosclerosis. Public Health Nutr. 2021;24(14):4630–41. [DOI] [PMC free article] [PubMed]
- 3.Bartlett JW, Harel O, Carpenter JR. Asymptotically unbiased estimation of exposure odds ratios in complete records logistic regression. Am J Epidemiol. 2015;182(8):730–6. [DOI] [PMC free article] [PubMed]
- 4.Besser LM, Chang LC, Hirsch JA, Rodriguez DA, Renne J, Rapp SR, et al. Longitudinal associations between the neighborhood built environment and cognition in US older adults: the multi-ethnic study of atherosclerosis. Int J Environ Res Public Health. 2021;18(15):7973. [DOI] [PMC free article] [PubMed]
- 5.Burstein R, Henry NJ, Collison ML, Marczak LB, Sligar A, Watson S, Filip I. Mapping 123 million neonatal, infant and child deaths between 2000 and 2017. Nature. 2019;574(7778):353–8. [DOI] [PMC free article] [PubMed]
- 6.Chen J, Kan K, Davis DS. Administrative reclassification and neighborhood governance in urbanizing China. Cities. 2021;118:103386.
- 7.De Stavola BL, Daniel RM, Ploubidis GB, Micali N. Mediation analysis with intermediate confounding: structural equation modeling viewed through the causal inference lens. Am J Epidemiol. 2015;181(1):64–80. [DOI] [PMC free article] [PubMed]
- 8.Du X, Jiao F. How the rural infrastructure construction drives rural economic development through rural living environment governance—case study of 285 cities in China. Front Environ Sci. 2023;11:1280744.
- 9.Feingold A, MacKinnon DP, Capaldi DM. Mediation analysis with binary outcomes: direct and indirect effects of pro-alcohol influences on alcohol use disorders. Addict Behav. 2019;94:26–35. [DOI] [PMC free article] [PubMed]
- 10.Frank LD, Iroz-Elardo N, MacLeod KE, Hong A. Pathways from built environment to health: a conceptual framework linking behavior and exposure-based impacts. J Transp Health. 2019;12:319–35.
- 11.Grocke-Dewey M, Hardison-Moody A, Haynes-Maslow L, Maras S, Webber E, Andress L, et al. Examining the relationship between physical activity and mental health during the COVID-19 pandemic across five US states. Prev Med Rep. 2021;24:101537. [DOI] [PMC free article] [PubMed]
- 12.Gong Y, Gallacher J, Palmer S, Fone D. Neighbourhood green space, physical function and participation in physical activities among elderly men: the Caerphilly prospective study. Int J Behav Nutr Phys Act. 2014;11(1):40. [DOI] [PMC free article] [PubMed]
- 13.Guo Q, Bai X, Feng N. Social participation and depressive symptoms among Chinese older adults: a study on rural–urban differences. J Affect Disord. 2018;239:124–30. [DOI] [PubMed]
- 14.Guo Y, Liu Y, Lu S, Chan OF, Chui CHK, Lum TYS. Objective and perceived built environment, sense of community, and mental wellbeing in older adults in Hong Kong: a multilevel structural equation study. Landsc Urban Plan. 2021;209:104058.
- 15.Hussain B, Mirza M, Baines R, Burns L, Stevens S, Asthana S, et al. Loneliness and social networks of older adults in rural communities: a narrative synthesis systematic review. Front Public Health. 2023;11:1113864. [DOI] [PMC free article] [PubMed]
- 16.Hu L, Wang L, Zhang Y, Wang K, Wang Y, Tan H, et al. Participation in household physical activity lowers mortality risk in Chinese women and men. Int J Environ Res Public Health. 2023;20(2):987. [DOI] [PMC free article] [PubMed]
- 17.Huang T, Larsen KT, Ried-Larsen M, Møller NC, Andersen LB. The effects of physical activity and exercise on brain-derived neurotrophic factor in healthy humans: a review. Scand J Med Sci Sports. 2014;24(1):1–10. [DOI] [PubMed]
- 18.Hunter S, Battista K, Leatherdale ST, Spence JC, Carson V. Longitudinal associations between the school-built environment and adolescents’ physical activity: evidence from the COMPASS study. J Transp Health. 2023;29:101565.
- 19.Jeong TW, Lee JW, Kim OJ, Choi JI, Kim HW, Choi YD. The effects of intensity of physical activity on mental health. J Med Imaging Health Inform. 2020;10(6):1413–7.
- 20.Kessler RC, Barker PR, Colpe LJ, Epstein JF, Gfroerer JC, Hiripi E, et al. Screening for serious mental illness in the general population. Arch Gen Psychiatry. 2003;60(2):184. [DOI] [PubMed]
- 21.Kim JI, Yu CY, Woo A. The impacts of visual street environments on obesity: the mediating role of walking behaviors. J Transp Geogr. 2023;109:103593.
- 22.Kong ST, Wu Q. Chinese family and society dynamics using the China family panel studies (CFPS) household panel. Aust Econ Rev. 2019;52(1):127–33.
- 23.Lan F, Pan J, Zhou Y, Huang X. Impact of the built environment on residents’ health: evidence from the China labor dynamics survey in 2016. J Environ Public Health. 2023;2023(1):3414849. [DOI] [PMC free article] [PubMed]
- 24.Lee ACK, Jordan HC, Horsley J. Value of urban green spaces in promoting healthy living and wellbeing: prospects for planning. Risk Manag Healthc Policy. 2015. 10.2147/RMHP.S61654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Lee SA, Ju YJ, Lee JE, Hyun IS, Nam JY, Han KT, et al. The relationship between sports facility accessibility and physical activity among Korean adults. BMC Public Health. 2016;16(1):893. [DOI] [PMC free article] [PubMed]
- 26.Lee S, Tsang A, Ng KL, Ma YL, Guo W, Mak A, et al. Performance of the 6-item Kessler scale for measuring serious mental illness in Hong Kong. Compr Psychiatry. 2012;53(5):584–92. [DOI] [PubMed]
- 27.Li A, Mansour A, Bentley R. Green and blue spaces, COVID-19 lockdowns, and mental health: an Australian population-based longitudinal analysis. Health Place. 2023;83:103103. [DOI] [PubMed]
- 28.Li J, Hirsch JA, Michael YL, Besser LM, Auchincloss AH, Hughes TM, et al. Spatial scale effects on associations between built environment and cognitive function: multi-ethnic study of atherosclerosis. Health Place. 2024;86:103181. [DOI] [PMC free article] [PubMed]
- 29.Little RJ, Carpenter JR, Lee KJ. A comparison of three popular methods for handling missing data: complete-case analysis, inverse probability weighting, and multiple imputation. Sociol Methods Res. 2024;53(3):1105–35.
- 30.Liu J, Yin C, Sun B. Associations between neighborhood environments and health status among Chinese older people during the pandemic: exploring mediation effects of physical activity. J Transp Health. 2024;35:101757.
- 31.Liu W, Zheng R, Zhang Y, Zhang W. Differences in the influence of daily behavior on health among older adults in urban and rural areas: evidence from China. Front Public Health. 2023;11:1259204. [DOI] [PMC free article] [PubMed]
- 32.Liu Y, Li J. Longitudinal associations between neighborhood characteristics and adults’ mental health in China: do age and urbanicity moderate the associations? Cities. 2024;152:105160.
- 33.Liu Y, Li J, Yang W. Nonlinear interactive associations between built environments and adults’ mental health in China using a longitudinal design. Appl Geogr. 2025;180:103662.
- 34.Liu Y, Lu S, Chen Y. Spatio-temporal change of urban–rural equalized development patterns in China and its driving factors. J Rural Stud. 2013;32:320–30.
- 35.Liu Z, Kemperman A, Timmermans H, Yang D. Heterogeneity in physical activity participation of older adults: a latent class analysis. J Transp Geogr. 2021;92:102999.
- 36.Markevych I, Schoierer J, Hartig T, Chudnovsky A, Hystad P, Dzhambov AM, et al. Exploring pathways linking greenspace to health: theoretical and methodological guidance. Environ Res. 2017;158:301–17. [DOI] [PubMed]
- 37.McKee AC, Daneshvar DH, Alvarez VE, Stein TD. The neuropathology of sport. Acta Neuropathol. 2014;127(1):29–51. [DOI] [PMC free article] [PubMed]
- 38.Moore TH, Kesten JM, López-López JA, Ijaz S, McAleenan A, Richards A, et al. The effects of changes to the built environment on the mental health and well-being of adults: systematic review. Health Place. 2018;53:237–57. [DOI] [PubMed]
- 39.Murrin E, Taylor N, Peralta L, Dudley D, Cotton W, White RL. Does physical activity mediate the associations between blue space and mental health? A cross-sectional study in Australia. BMC Public Health. 2023;23(1):203. [DOI] [PMC free article] [PubMed]
- 40.Muthén LK, Muthén BO. Mplus: statistical analysis with latent variables: user’s guide version 8.3. Los Angeles, CA: Authors; 2019.
- 41.Nybo L, Secher NH. Cerebral perturbations provoked by prolonged exercise. Prog Neurobiol. 2004;72(4):223–61. [DOI] [PubMed]
- 42.Pasanen TP, White MP, Wheeler BW, Garrett JK, Elliott LR. Neighbourhood blue space, health and wellbeing: the mediating role of different types of physical activity. Environ Int. 2019;131:105016. [DOI] [PubMed]
- 43.Pelletier C, White N, Duchesne A, Sluggett L. Work, travel, or leisure: comparing domain-specific physical activity patterns based on rural–urban location in Canada. BMC Public Health. 2023;23(1):2216. [DOI] [PMC free article] [PubMed]
- 44.Putra IGNE, McInerney AM, Robinson E, Deschênes SS. Neighbourhood characteristics and socioeconomic inequalities in child mental health: cross-sectional and longitudinal findings from the growing up in Ireland study. Health Place. 2024;86:103180. [DOI] [PubMed]
- 45.Rappaport LM, Amstadter AB, Neale MC. Model fit estimation for multilevel structural equation models. Struct Equ Modeling. 2020;27(2):318–29. [DOI] [PMC free article] [PubMed]
- 46.Rhodes RE, Zhang R, Zhang CQ. Direct and indirect relationships between the built environment and individual-level perceptions of physical activity: a systematic review. Ann Behav Med. 2020;54(7):495–509. [DOI] [PubMed]
- 47.Richardson EA, Pearce J, Mitchell R, Kingham S. Role of physical activity in the relationship between urban green space and health. Public Health. 2013;127(4):318–24. [DOI] [PubMed]
- 48.Sallis JF, Cerin E, Conway TL, Adams MA, Frank LD, Pratt M, et al. Physical activity in relation to urban environments in 14 cities worldwide: a cross-sectional study. Lancet. 2016;387(10034):2207–17. [DOI] [PMC free article] [PubMed]
- 49. SNPC. 2010. The Sixth National Population Census. https://www.stats.gov.cn/sj/pcsj/rkpc/6rp/indexch.htm. Accessed 21 Oct 2023.
- 50.Solbraa AK, Anderssen SA, Holme IM, Kolle E, Hansen BH, Ashe MC. The built environment correlates of objectively measured physical activity in Norwegian adults: a cross-sectional study. J Sport Health Sci. 2018;7(1):19–26. [DOI] [PMC free article] [PubMed]
- 51.Sugiyama T, Leslie E, Giles-Corti B, Owen N. Associations of neighbourhood greenness with physical and mental health: do walking, social coherence and local social interaction explain the relationships? J Epidemiol Community Health. 2008;62(5):e9–e9. [DOI] [PubMed]
- 52.Sui Y, Ettema D, Helbich M. Longitudinal associations between the neighborhood social, natural, and built environment and mental health: a systematic review with meta-analyses. Health Place. 2022;77:102893. [DOI] [PubMed]
- 53.Tang D, Bu T, Feng Q, Liu Y, Dong X. Differences in overweight and obesity between the north and south of China. Am J Health Behav. 2020;44(6):780–93. [DOI] [PubMed]
- 54. The State Council of the People's Republic of China. (1978). Provisional measures of the state council on the retirement and retirement of workershttps://rsj.dl.gov.cn/old/dlrsw/sites/default/files/ueditor/16/upload/file/20201028/1603873621139415.pdf. Accessed 26 Oct 2020.
- 55.Wang M, Huang Y, Song Y, Chen J, Liu X. Study on environmental and lifestyle factors for the North-South differential of cardiovascular disease in China. Front Public Health. 2021;9:615152. [DOI] [PMC free article] [PubMed]
- 56.Wedyan M, Saeidi-Rizi F. Assessing the impact of urban environments on mental health and perception using deep learning: a review and text mining analysis. J Urban Health. 2024;101(2):327–43. [DOI] [PMC free article] [PubMed]
- 57.White IR, Daniel R, Royston P. Avoiding bias due to perfect prediction in multiple imputation of incomplete categorical variables. Comput Stat Data Anal. 2010;54(10):2267–75. [DOI] [PMC free article] [PubMed]
- 58.Wittchen HU. Screening for serious mental illness: methodological studies of the K6 screening scale. Int J Methods Psychiatr Res. 2010;19(S1):1–3. [DOI] [PMC free article] [PubMed]
- 59.Xu X, Lei Z, Dou Y, Liu S. Research on gap of balanced development between the North and the South of China—analysis based on “China Balanced Development Index.” China Industrial Economics. 2021;2:5–22 ((In Chinese)). [Google Scholar]
- 60.Yang W, Li Y, Liu Y, Fan P, Yue W. Environmental factors for outdoor jogging in Beijing: insights from using explainable spatial machine learning and massive trajectory data. Landsc Urban Plann. 2024;243:104969.
- 61.Yang W, Yang R, Li X. A canonical correlation analysis study on the association between neighborhood green space and residents’ mental health. J Urban Health. 2023;100(4):696–710. [DOI] [PMC free article] [PubMed]
- 62.Yang Y, Bao W, Wang Y, Liu Y. Measurement of urban-rural integration level and its spatial differentiation in China in the new century. Habitat Int. 2021;117:102420.
- 63.Yue Y, Yang D, Owen N, Van Dyck D. The built environment and mental health among older adults in Dalian: the mediating role of perceived environmental attributes. Soc Sci Med. 2022;311:115333. [DOI] [PubMed]
- 64.Zhang CJ, Barnett A, Sit CH, Lai PC, Johnston JM, Lee RS, et al. Cross-sectional associations of objectively assessed neighbourhood attributes with depressive symptoms in older adults of an ultra-dense urban environment: the Hong Kong ALECS study. BMJ Open. 2018;8(3):e020480. [DOI] [PMC free article] [PubMed]
- 65.Zhang R, Zhang CQ, Rhodes RE. The pathways linking objectively-measured greenspace exposure and mental health: a systematic review of observational studies. Environ Res. 2021;198:111233. [DOI] [PubMed]
- 66.Zhang T, Chiu RLH, Ho HC. Suburban neighborhood environments and depression: a case study of Guangzhou, China. J Transp Health. 2019;15:100624.
- 67.Zhang Y, Chai Y. Study on suburbanization of living and activity space. Prog Geogr. 2013;32(12):1723–31.
- 68.Zhu W, Chi A, Sun Y. Physical activity among older Chinese adults living in urban and rural areas: a review. J Sport Health Sci. 2016;5(3):281–6. [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
(PDF 259 KB)