Buffering effects of public sports facilities on physical activity and health equity during commercial facility fluctuations: A natural experiment in Shenzhen, China

Buffering effects of public sports facilities on physical activity and health equity during commercial facility fluctuations: A natural experiment in Shenzhen, China | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Buffering effects of public sports facilities on physical activity and health equity during commercial facility fluctuations: A natural experiment in Shenzhen, China Jingyi Yang, Peiling Zhou This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6484311/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background accessibility to physical activity (PA) facility can strongly influence population health outcomes through PA engagement duration in neighborhoods. While commercial facilities tend to spatially cluster, creating inequitable PA opportunities and health disparities between neighborhoods, little is known about how public facilities buffer against such neighborhood health inequities. Regulation of facility openings during the COVID-19 pandemic provided a natural experimental setting to examine this relationship. Methods We conducted a longitudinal study of PA behaviors among 701 residents from 23 neighborhoods in Shenzhen during facility closure and reopening (2019–2020). Using difference-in-differences (DID) analysis, we examined how different types of facilities influenced various PA duration. Through triple-difference (DDD) analysis, we investigated how public facilities moderated PA duration responses to commercial facility changes within 1,000-meter catchments. Results PA facilities showed distinct functional effects, with parks increasing light physical activity (LPA) ( β = 39.45, p = 0.032) and public sports facilities enhancing moderate-to-vigorous physical activity (MVPA) ( β = 39.07, p = 0.015); Commercial facilities exhibited spatial polarization with diminishing marginal returns, showing strong effects on MVPA in low-accessibility areas ( β = 70.02, p < 0.001) but negligible impact in high-accessibility areas ; Through triple-difference analysis, we quantified the conditional buffering effect where public sports facilities effectively mitigated MVPA reduction in areas experiencing commercial facility disadvantages ( β = 79.27, p = 0.050), with this effect being strongest in commercially deprived areas ( β = 42.51, p = 0.001). Conclusions Through spatially balanced distribution and complementary functional design, public sports facilities effectively mitigated neighborhood MVPA inequities caused by commercial facility clustering. This conditional buffering effect was particularly significant in commercially disadvantaged areas, informing public health and urban planning policies for building resilient and equitable urban PA environments. Sport facility accessibility Built environment Physical activity Health inequity Neighborhood deprivation Natural experiment Public health policy Figures Figure 1 Figure 2 Introduction Extensive research has demonstrated the negative impact of physical inactivity on non-communicable diseases, emphasizing the importance of boosting physical activity levels across populations to enhance public health(Kohl et al., 2012 ; Li et al., 2024 ). According to the socioecological model, environmental factors play a crucial role affecting PA behaviors(Stokols, 1996 ). Facilities at neighborhood serve as primary venues for residents' PA engagement, and their accessibility is strongly associated with population health status(Salvo et al., 2018 ). Comprehensive PA infrastructure, such as sports venues, fitness rooms, and public green Spaces, creates opportunities for PA participation(Durão et al., 2023 ), thereby enhancing physical fitness, preventing non-communicable diseases, and improving mental health outcomes(Gomes et al., 2023 ; Menardo et al., 2022 ; White et al., 2024 ). Neighborhood deprivation, including PA facility scarcity in some neighborhood contributes to health inequities(Kegler et al., 2022 ; Rocha et al., 2017 ; Sawyer et al., 2017 ). Residents in communities with low accessibility to PA facilites exhibit lower PA levels and ambulatory activities compared to neighborhoods with high sports facility accessibility(Eriksson et al., 2012 ; van Lenthe et al., 2005 ), and thus have poorer health outcomes(Lee et al., 2022 ; Reijneveld et al., 2000 ). More recent models emphasize the combined impacts between environmental and social factors upon health outcomes(Olvera Alvarez et al., 2018 ). For example, public and commercial PA facilities in neoliberal cities shows divergent patterns of spatial distribution and serve for different groups of populations (O’Brien, 2021 ; Santacruz Lozano et al., 2021 ). Public PA facilities, most of which are government-subsidized and accessible to the general public, are designed to offer PA opportunities for a wider population. According to WHO Guidelines on Physical Activity and Sedentary Behavior(2020), governments in many countries employ these facilities as interventions to promote PA duration, although their availability is typically limited by resource allocation constraints (Bergsgard et al., 2019 ; Kang & Lee, 2022 ). In contrast, commercial facilities may cause inequal and even bipolar health outcomes between populations of different socioeconomic status. Funded and operated by private entities and accessible only to members, commercial facilities tend to offer specialized and diverse sports programs. These facilities primarily cater to individuals willing to pay for particular services (Kang & Lee, 2022 ; Rivera-Navarro et al., 2022 ) and are often concentrated in areas of higher socioeconomic status due to market dynamics (Lisbona Guillén et al., 2024 ). In China, while the central and local governments have taken multiple measures to construct PA facilities in urban neighborhoods to intervene physical inactivity and to promote public health, limited financial expenditure on health intervention and the marketization of public health industry cause spatial inequities in public and commercial PA facility accessibility at different geographic scales, across regions(Jiang et al., 2023 ), between urban and rural areas (Jing et al., 2018 ), across and within cities (Xu et al., 2022 ) and districts(Yang & Liu, 2024 ). These multidimensional aspects of neighborhood deprivation contribute to health inequities through disparate access to PA opportunities(Jiang et al., 2023 ; Y. Liu et al., 2020 ) . While the uneven distribution of public and commercial PA facilities are simultaneously causing health inequalities, little research has simultaneously investigated the integrated effect of both public and commercial PA facilities(Macdonald, 2019 ) nor fully elucidated the causal effects of varying sports facility accessibility on PA behaviors (Sawyer et al., 2017 ). The closure and subsequent reopening of sports facilities during the COVID-19 pandemic (Hoekman et al., 2023 ; Lee et al., 2022 ) present an ideal natural experimental setting to investigate the relationship between sports facility accessibility and PA levels (Ding et al., 2020 ). This study examines the neighborhood-level PA facilities and their impacts on residential PA patterns to provide evidence-based recommendations for equitable sports facility planning in urban environments. In particular, the objectives and hypotheses of this study include: Objective 1 To examine how inequitable distribution of neighborhood sports facilities affects residents' PA duration disparities Hypothesis 1 The spatial inequality in sports facility distribution creates persistent differences in neighborhood PA duration. Areas with higher facility accessibility demonstrate greater PA duration and faster recovery following facility reopening, with parks primarily facilitating LPA and sports venues promoting MVPA. Objective 2 To investigate how changes in commercial sports facility accessibility polarize residents' PA duration Hypothesis 2 Changes in commercial facility accessibility create variations in MVPA duration. This effect is particularly pronounced in areas with limited commercial facility accessibility but negligible in areas with adequate commercial sports resources. Objective 3 To evaluate how public facilities buffer against PA duration inequities caused by spatial clustering of commercial facilities Hypothesis 3 In areas with limited commercial facility accessibility, public sports facilities mitigate the impact of commercial facility fluctuations on residents' PA duration by providing alternative venues particularly in communities experiencing reductions or scarcity in commercial facilities. Methods Research Design This study employed a natural experimental design, utilizing operational changes in commercial sports facilities during the COVID-19 pandemic as a quasi-natural intervention. We conducted a retrospective longitudinal assessment in November 2020 (T2), collecting both current data and recalled baseline data from November 2019 (T1). A difference-in-differences (DID) approach was applied to evaluate the impact of changes in facility accessibility on PA behavior, allowing us to identify intervention effects while controlling for temporal trends and unobserved confounders. The temporal framework encompassed: (1) baseline period (T1) in November 2019 when facilities operated normally; (2) facility closure phase beginning in late January 2020 due to COVID-19 containment measures; (3) differential reopening phase from March 2020, with public facilities gradually resuming operations while some commercial facilities permanently closed; and (4) Stable new facility landscape with full restoration of public facilities in November 2020 (T2). The matching seasons (November) in baseline and follow-up measurements helped control for seasonal variation in PA duration. Figure 1 illustrates our natural experiment framework, showing the baseline and follow-up conditions, the intervention point created by the pandemic, and assess how changes in facility accessibility affected physical activity patterns. Study area and sampling procedures A multi-stage sampling strategy was employed to recruit participants from 23 residential communities across six administrative districts in Shenzhen, China, ensuring geographic and socioeconomic representation. Participants lived in a variety of housing types including commercial housing, urban villages, danwei compounds, and public housing. we recruited a minimum of 30 eligible participants aged 18–65 years from each community. A total of 701 participants provided valid responses in November 2020, including their current PA duration per week and those retrospectively reported in November 2019. The sample composition reflected the distribution of sex, age, and occupation. All participants provided written informed consent, and the study protocol was approved by the Ethics Committee of the University that the first author is studying at. Data Collection Physical Activity Data Collection Protocol PA data were collected using the Day Reconstruction Method (DRM), following protocols detailed in our previous work(Kahneman et al., 2004 ; Zhou et al., 2024 ).Trained interviewers conducted standardized face-to-face interviews between November and December 2020, where participants recalled their typical weekday activity patterns for November 2019 (T1) and November 2020 (T2). To minimize recall bias, participants verified their activity information against smartphone payment records and mobility trajectories. Environmental Facility Data Environmental Facility Data were collected from multiple authoritative sources. Public facilities records, including parks and sports venues, were obtained from the website of Culture, Radio, Television, Tourism and Sports Bureau of Shenzhen Municipality and Urban Management And Law Enforcement (2019). Commercial sports facility records were extracted from Gaode Maps Points of Interest (POI) database at two time points: baseline (T1: November 2019) and follow-up (T2: November 2020) and were triangulated with facility utilization records from mobile phone data. All facility locations were geocoded using a standardized protocol to calculate spatial accessibility indices. Measurements Physical Activity Classification and Assessment Weekly PA duration served as the primary outcome measure. Activities were categorized based on metabolic equivalents (METs) into moderate-to-vigorous physical activity(MVPA)(METs ≥ 3.0, such as gym workouts, ball sports, and running) and light physical activity(LPA) (1.5 < METs < 3.0, such as walking and Walk the dog)(Ainsworth, 2024 ). Participants reported activity type, location, weekly frequency, and duration per session across three daily periods (morning, noon, evening). Weekly PA duration per week was calculated by multiplying frequency with per-session duration. From 921 initially collected activity records, 841 valid records from 701 participants were retained after excluding samples missing critical information, or falling without the neighborhood scope. Accessibility of Sports Facilities Following Gidlow (2019)and M. Liu et al ( 2024 ), the accessibility of PA facilities was measured by the number of facilities within 1,000-meter residential catchment areas with China's urban planning policy for "15-minute community life circles". Environmental intervention indicators were defined based on facility density changes during COVID-19. We measured the number of three types of facilities within the 1000m residential buffer, and defined the high-low level of facilities according to the median values of each density. Two types of public facilities were considered: (1) park accessibility: high-density areas defined as ≥ 4 facilities; (2) public sports facility accessibility: categorized by presence (≥ 1 facility).For commercial sports facilities, which served as the primary intervention variable, baseline high-density areas were defined as > 43 facilities, with subsequent change status classified as stable (reduction ≤ 10% from baseline) or decline (reduction > 10%), The 10% threshold was established through empirical analysis as a natural break in facility change distribution, effectively distinguishing between normal market fluctuations ( 10%) that significantly impact resident behaviors(Park et al., n.d.; Zhang et al., 2024 ) Covariates Our analysis incorporated two sets of covariates: individual characteristics and built environmental features. Individual variables included individual characteristic: age (categorized as 65 years), gender (male = 0, female = 1), educational attainment (high school or below = 0, bachelor's degree or above = 1), and annual household income. These categorizations align with previous studies examining associations between built environment and PA behaviors (Jowshan et al., 2025 ; Lundh et al., 2025 ). Built environmental-level characteristics were measured within participants' walking catchment areas using ArcGIS 10.8 (Esri, 2020 ). These measures included street connectivity, building density, and land-use mix. Statistical Analysis We employed quasi-experimental methods - Difference-in-Differences (DID) (Singer & Willett, 2003 )and Triple Differences (DDD)(Olden & Møen, 2022 ) approaches to evaluate how different levels of facility accessibility causally affect PA duration. These methods allow us to estimate the impact of facility accessibility changes by controlling for time-invariant unobserved heterogeneity and time-specific shocks. In all models: • Yit represents individual i's weekly PA duration at Time t , analyzed separately for: a) Light Physical Activity (LPA) Duration b) Moderate-to-Vigorous Physical Activity (MVPA) Duration c) Total Physical Activity Duration • Time i indicates pre (T1) and post (T2) periods • γXit controls for baseline effects(time, pub, com), built environment features (population density, land-use mix, street connectivity), and individual characteristics (age, gender, education, income) • αpark_j accounts for park-specific fixed effects • εit represents the error term Parallel Trend Assumption We assume that, in the absence of any facility-related interventions (including changes in facility numbers or operational status), the PA duration would follow similar trends across different areas over time. That is, if facility accessibility had no causal impact on PA duration, the change in PA duration should not differ across these areas before and after the intervention. The net effect was calculated as: impact_of_facility = (PA high_access,t2 - PA high_access,t1 ) - (PA low_access,t2 - PA low_access,t1 ) where PA is the outcome of physical activity in the control group (low accessibility areas) vs. the treatment group (high accessibility areas) at a particular time period (T1 or T2). We used a parametric model to obtain standard errors and significance levels for the DID estimate : DID Analysis: Primary Effects of Facility Accessibility First, we employed a general DID model to assess how different levels of facility accessibility (parks, public sports facilities, and commercial facilities) influenced PA duration. Given the distinct distribution patterns of facilities (park numbers: range 0–35, mean 5.23, SD 4.775; public facilities: range 0–4, median 1) and the few variations in park accessibility over the study period, we incorporated park numbers as fixed effects while classifying areas into high/low accessibility levels for public and commercial facilities at baseline. This specification demonstrated superior model fit. Yit = β0 + β1(Park i × Time t ) + β2(Pub i ×Time t ) + β3(Com i ×Time t ) + αpark_j + γXit + εit Where: • Park i , pub i and Com i indicate accessibility levels (high = 1, low = 0) for parks, public sports facilities, and commercial facilities respectively • β0 represents the baseline average • β1 - β3 are the DID estimators that measure the differential changes in PA over time between areas with varying facility accessibility levels DID Analysis: Commercial Facility Change Effects We then focused on how changes in commercial facility accessibility affected PA duration: Yit = β0 + β1 (Change i × Time t ) + αpark_j + γXit + εit Where Change i indicates the change of the commercial facilities (stable/increased = 1, decreased = 0). We conducted this analysis for both the full sample and subsamples stratified by original commercial facility accessibility to examine potential heterogeneous effects. DDD Analysis: Public Facility Buffer Effects We employed a DDD model to examine how public sports facilities moderated the relationship between commercial facility changes and PA duration, particularly focusing on areas with either low baseline accessibility or decreasing trends in commercial facilities: Y it = β0 + β1 (Pub i × Com_disadv i × Time t ) + β2 (Pub i × Time t ) + β3 (Com_disadv i × Time t ) + β4 (Pub i × Com i ) + αpark_j + γXit + εit Where: • Pub i indicates public sports facility accessibility (high = 1, low = 0) • Com_disadv i represents disadvantaged areas in terms of commercial facilities (high baseline accessibility and stable/increased trend = 0, others = 1) DID Analysis: Public Facility Buffer Effects in Different Commercial Facility Areas We further investigated the buffering role of public facilities in areas with different commercial facility contexts by stratifying the sample into two groups: areas with commercial facility disadvantage (Com_disadv i =1) and areas with commercial facility advantage (Com_disadv i =0). We then conducted a stratified analysis using the following model: Yit = β0 + β1 (Pub i × Time t ) + αpark_j + γXit + εit Where Pub i indicates public sports facility accessibility (high = 1, low = 0). This stratified analysis allowed us to assess whether the impact of public facilities on PA declines differed based on the pre-existing commercial facility environment and its changing trajectory during the pandemic. Model Validation The Robustness analysis was performed adjusting control variables to test the robustness of the results. All analyses were conducted using STATA 17.0 (StataCorp, 2021 ), Statistical significance was set at p < 0.05. We assessed multicollinearity among independent variables before regression analysis. Variance inflation factors (VIF) for all predictors remained below 5 (See Supplementary Table 1, Additional File 1), and Pearson correlation coefficients showed no values exceeding 0.7 (See Supplementary Table 2, Additional File 1), confirming acceptable independence among variables. Results Descriptive Statistics The study sample comprised 701 participants with a balanced sex distribution (50% each). The majority of participants (67%) were aged 26–45 years, with 69% holding a bachelor's degree or higher. Sixty percent reported annual household incomes between 120,000-360,000 CNY. The study areas were characterized by mean road network density of 0.025 (SD = 0.005), land-use mix index of 0.701 (SD = 0.063), and building density index of 0.205 (SD = 0.067). Detailed demographic characteristics and environmental indicators are presented in Table 1 . Spatial analysis revealed distinct distribution patterns of different facility types across neighborhoods. Commercial sports facilities, predominantly clustered in central urban areas with high socioeconomic status, experienced a substantial reduction from 2019 (Fig. 2 a) to 2020 (Fig. 2 b), with mean density decreasing from 70.194 (SD = 111.576) to 35.578 (SD = 21.857) facilities per catchment area. This change exhibited marked polarization: despite decreasing from 112to 45.8 facilities, high commercial facility accessibility areas maintained levels above the threshold of 43 facilities, while low commercial facility accessibility areas decreased from 31.033 to 25.983, consistently remaining below the threshold. In contrast, public sports facilities (Fig. 2 c and 2 d) demonstrated more balanced and stable spatial distribution, maintaining a mean density of 1.02 (SD = 1.171), supplemented by parks (Fig. 2 e and 2 f) with a density of 5.233 (SD = 4.775). This spatial configuration forms the foundation for examining how different facility types influenced residents' PA recovery patterns. Following the reopening of sports facilities, participants' PA duration showed varying recovery trends. Overall (See Supplementary Table 3, Additional File 1), mean weekly LPA increased from 73.172 to 84.711 minutes, MVPA from 57.803 to 68.266 minutes, and total PA duration from 130.975 to 152.977 minutes per week. Spatial analysis revealed differential recovery patterns based on facility accessibility. People who live in areas with high park accessibility (≥ 4 parks) demonstrated not only recovery but exceeded T1 LPA levels (from 80.797 to 116.799 minutes/week), while low park accessibility areas (< 4 parks) showed decreased levels (from 68.308 to 64.243 minutes/week). Similarly, areas with high public facility accessibility exhibited increased MVPA duration (from 60.656 to 88.929 minutes/week), whereas low-accessibility areas experienced a decline (from 54.265 to 42.652 minutes/week). Areas with stable commercial facilities showed enhanced MVPA levels (from 62.535 to 93.655 minutes/week), while areas with decreased facilities remained below T1 levels (from 54.693 to 51.581 minutes/week). following commercial facility changes (T2), areas with high public facility accessibility and advantage commercial facilities maintained their elevated MVPA levels (from 86.896 to 88.63 minutes/week). Conversely, areas with both low public facility accessibility and disadvantaged accessibility to commercial facilities (n = 81) experienced a further decrease in MVPA, from 54.046 to 34.282 minutes/week. Table 1 Sample Characteristics and Environmental Features Characteristic n (%) or Mean(SD) Demographic characteristics Age (years) 18–25 132 (19.0%) 26–35 261 (37.0%) 36–45 213 (30.0%) 46–55 82 (12.0%) ≥ 56 13 (2.0%) Sex Male 349 (50.0%) Female 352 (50.0%) Educational attainment Below bachelor's degree 219 (31.0%) Bachelor's degree or above 482 (69.0%) Annual household income (CNY) 600,000 62 (9.0%) Environmental characteristics Road network density 0.025 (0.005) Land-use mix index 0.701 (0.063) Building density index 0.205 (0.067) Park density 5.233 (4.775) Public sports facility density 1.02 (1.171) Commercial sports facility density Total T1 70.194 (111.576) T2 35.578 (21.857) High accessibility T1 112.012 (149.307) T2 45.823 (24.58) Low accessibility T1 31.033 (9.352) T2 25.983 (13.015) Effects of Facilities Accessibility on PA duration Our difference-in-differences (DID) analyses revealed differential effects of facility accessibility on physical activity patterns. As shown in Table 2 , high park accessibility increased LPA duration ( β = 39.448, p = 0.032) and total PA duration ( β = 53.747, p = 0.020). Areas with high accessibility to public sports facilities showed enhanced MVPA duration ( β = 39.069, p = 0.015) and increased total physical activity duration ( β = 48.791, p = 0.032). However, commercial facility accessibility did not significantly influence PA duration. Among built environment characteristics, building density showed a positive association with total physical activity duration ( β = 301.272, p = 0.015), while land-use mix index was negatively associated with MVPA duration ( β =-128.333, p = 0.061). Regarding individual characteristics factors, age showed a positive association with LPA duration ( β = 26.867, p < 0.001) but a negative association with MVPA duration ( β =-11.734, p = 0.008). Additionally, age was positively associated with total PA duration ( β = 15.134, p = 0.016). Compared to males, females demonstrated lower MVPA duration ( β =-31.985, p < 0.001). Higher educational attainment was positively associated with MVPA duration ( β = 24.674, p = 0.011), while higher annual household income was associated with increased LPA duration ( β = 7.338, p = 0.028). The non-significant coefficients for both public sports facilities (LPA: β =-14.393, p = 0.297; MVPA: β =-5.133, p = 0.677; Total PA: β =-19.526, p = 0.263) and commercial sports facilities (LPA: β = 6.909, p = 0.621; MVPA: β =-8.279, p = 0.507; Total PA: β =-1.369, p = 0.938) support the parallel trends assumption, suggesting that pre-intervention physical activity patterns were similar across areas with different facility accessibility. Table 2 Effects of facility accessibility on PA duration: DID analysis Variables LPA duration MVPA duration Total PA duration β(P-value) β (P-value) β (P-value) DID ( Park × Time ) 39.448(0.032*) 14.299(0.382) 53.747(0.020*) DID ( Pub × Time ) 9.721(0.590) 39.069(0.015*) 48.791(0.032*) DID ( com × Time ) -8.843(0.621) 3.759(0.814) -5.084(0.822) Public facilities -14.393(0.297) -5.133(0.677) -19.526(0.263) Commercial facilities 6.909(0.621) -8.279(0.507) -1.369(0.938) Time -4.915(0.778) -18.553(0.235) -23.468(0.288) Road network density -2242.578(0.131) -365.269(0.783) -2607.846(0.164) Building density index 166.041(0.089) 135.230(0.121) 301.272(0.015**) Land-use mix index 8.143(0.915) -128.333(0.061*) -120.190(0.214) Educational attainment -18.764(0.083*) 24.674(0.011**) 5.909(0.665) Sex -11.357(0.211) -31.985(0.000***) -43.342(0.000***) Age 26.867(0.000***) -11.734(0.008***) 15.134(0.016**) Annual household income 7.338(0.028*) 0.556(0.852) 7.891(0.061*) Constant 46.285(0.482) 137.136(0.020*) 183.421(0.028*) R² 0.083 0.0792 0.0807 Observations 1402 1402 1402 p < 0.05, ** p < 0.01, *** p < 0.001. N = 1,402 observations (701 individuals × 2 time points) Effects of Commercial Sports Facilities Accessibility Changes on PA duration: As shown in Table 3 , in the full sample, the stability of commercial facilities positively influenced both MVPA duration ( β = 41.033, p = 0.001) and total PA duration ( β = 38.181, p = 0.032). These effects exhibited distinct patterns across neighborhoods with different baseline accessibility levels. The impact was particularly pronounced in low commercial facility accessibility neighborhoods, which is attributed to the stability of commercial facilities leading to a substantial increase in MVPA duration ( β = 70.019, p = 0.000) and total PA duration ( β = 77.406, p = 0.004). In contrast, high commercial facility accessibility neighborhoods showed minimal response to commercial facility changes (MVPA: β = 19.708, p = 0.252; total PA: β =-0.179, p = 0.994), suggesting a diminishing marginal effect of additional commercial facilities in areas with adequate existing resources. Table 3 Effects of Commercial Facility Changes on PA duration: Stratified DID Analysis Panel A: Full Sample Analysis Variables LPA duration MVPA duration Total PA duration β(P-value) β (P-value) β (P-value) DID( Change i × Time t ) -2.853(0.839) 41.033(0.001***) 38.181(0.032*) Public facilities -9.359(0.372) 11.899(0.203) 2.541(0.848) Commercial facilities 2.112(0.846) -0.991(0.919) 1.121(0.935) Time 12.670(0.228) 5.809(0.535) 6.861(0.606) Control Variables Yes Yes Yes R² 0.0794 0.0816 0.0764 Observations 1,402 1,402 1,402 Panel B: Stratified Analysis by Baseline Accessibility Low commercial facility accessibility DID ( Change i × Time t ) 7.387(0.723) 70.019(0.000***) 77.406(0.004***) Public facilities 20.039(0.308) 3.363(0.847) 23.402(0.349) Time 12.221(0.464) -24.188(0.103) -11.966(0.573) Control Variables Yes Yes Yes R² 0.0779 0.1247 0.1028 Observations 722 722 722 High commercial facility accessibility DID ( Change i × Time t ) -19.886(0.314) 19.708(0.252) -0.179(0.994) Public facilities -39.415(0.006**) 22.542(0.070*) -16.873(0.338) Time 12.874(0.328) 4.950(0.665) 17.825(0.272) Control Variables Yes Yes Yes R² 0.1401 0.1293 0.1259 Observations 680 680 680 Note: Control Variables include built environment and individual characteristics. p < 0.05, ** p < 0.01, *** p < 0.001. Control Variables include built environment features (population density, land-use mix, street connectivity), individual characteristics (age, gender, education, income), and park-specific fixed effects This polarizing effect was further evidenced by neighborhood-specific built environment characteristics. In high-accessibility neighborhoods, the influence of commercial facility changes was overshadowed by existing built environment features, particularly building density (total PA: β = 719.939, p = 0.000). These findings confirm our hypothesis that the effectiveness of commercial facility changes is contingent upon the baseline accessibility level, with areas of limited commercial resources benefiting most from such interventions. Joint Effects of Commercial and Public Sports Facilities on PA duration: The DDD analysis revealed patterns in how public sports facilities moderate physical activity (PA) inequities. As shown in Table 4 , the triple-interaction term showed a positive association with MVPA duration ( β = 79.269, p = 0.050), indicating that public facilities effectively buffer against MVPA inequality in commercially disadvantaged areas. This positive coefficient suggests that public facilities have stronger positive effects on MVPA in neighborhoods with commercial facility disadvantages (either low baseline accessibility or decreasing trends) compared to commercially advantaged areas. Notably, no significant interaction between facility types was observed prior to facility changes ( β =-31.337, p = 0.168), suggesting that the equity-promoting effect of public facilities emerged specifically in response to commercial facility changes. The effectiveness of this buffering mechanism was further contextualized by both built environment and individual characteristics. Among built environment factors, building density showed positive associations with total PA duration ( β = 305.468, p = 0.014). Individual characteristics factors also played significant roles, with educational attainment ( β = 23.599, p = 0.015) showing a positive association, while both female gender ( β =-31.200, p < 0.001) and age ( β =-11.878, p = 0.008) were negatively associated with MVPA duration. Annual household income showed a positive association with LPA duration ( β = 6.991, p = 0.037) but not with MVPA specifically ( β = 0.897, p = 0.763). Table 4 Joint Effects of Public and Commercial Facilities: Triple-Difference Analysis Variables LPA duration MVPA duration Total PA duration β(P-value) β (P-value) β (P-value) DDD ( Pub i × Com_disadv i × Time t ) -19.575(0.667) 79.269(0.050*) 59.694(0.299) Pub i × Time t 31.449(0.464) -28.546(0.456) 2.902(0.957) Com_disadv i × Time t 34.905(0.269) -47.683(0.009**) -12.777(0.749) Pub i × Com_disadv i 10.510(0.680) -31.337(0.168) -20.827(0.518) Public facilities -24.545(0.326) 20.264(0.363) -4.281(0.892) Commercial facilities 7.105(0.538) -10.738(0.296) 3.632(0.803) Time -26.690(0.386) 23.280(0.269) 3.590(0.926) Road network density -1957.725(0.192) -714.726(0.592) -2672.448(0.158) Building density index 160.960(0.101) 144.508(0.099) 305.468(0.014**) Land-use mix index -10.442(0.896) -112.880(0.111) -123.322(0.220) Educational attainment -17.748(0.102) 23.599(0.015**) 5.852(0.669) Sex -12.028(0.187) -31.200(0.000***) -43.228(0.000***) Age 27.038(0.000***) -11.878(0.008***) 15.160(0.016**) Annual Household Income 6.991(0.037*) 0.897(0.763) 7.884(0.062*) Constant 53.046(0.424) 135.660(0.022*) 188.707(0.025*) R² 0.0813 0.0818 0.078 Observations 1402 1402 1402 p < 0.05, ** p < 0.01, *** p < 0.001 Further analysis stratified by commercial facility advantage (Table 5 ) revealed that while public facilities positively promoted MVPA in both commercially advantaged and disadvantaged areas, the effect was significant only in commercially disadvantaged neighborhoods ( β = 42.506, p = 0.001). In contrast, commercially advantaged areas showed a positive but non-significant effect ( β = 6.464, p = 0.868). This finding suggests that public facilities play a crucial role in mitigating MVPA inequalities, particularly in areas experiencing commercial facility disadvantages. Table 5 Effects of Public Sports Facilities on PA Duration by Commercial Facility Advantage Disadvantage Commercial Facilities Advantage Commercial Facilities (n = 601) (n = 100) Variables LPA duration MVPA duration Total PA duration LPA duration MVPA duration Total PA duration DID( Change × Time t -3.261(0.827) 42.506(0.001***) 39.245(0.034*) -13.372(0.677) 6.464(0.868) -6.909(0.886) Time 16.267(0.202) -14.026(0.200) 2.242(0.887) 2.041(0.941) 15.067(0.648) 17.108(0.677) Control Variables Yes Yes Yes Yes Yes Yes R² 0.0752 0.0798 0.0827 0.3085 0.1851 0.2103 p < 0.05, ** p < 0.01, *** p < 0.001. Control Variables include built environment features (population density, land-use mix, street connectivity), individual characteristics (age, gender, education, income), and park-specific fixed effects These findings provide empirical evidence supporting the strategic placement of public sports facilities as an intervention tool to promote PA equity, particularly in areas experiencing commercial facility disadvantages. Robustness tests The robustness checks of our findings through progressive addition of control variables demonstrated consistent results (Wang et al., 2024 ). As shown in Supplementary Table 4, Additional File 1, our results demonstrate strong stability across different model specifications. Specifically, park accessibility shows consistent positive effects on LPA (39.45–51.86 minutes/week, p < 0.05). While public sports facilities show no significant impact on LPA, they demonstrate stable positive effects on MVPA (31.17–40.94 minutes/week, p < 0.01) and total physical activity (26.81–48.79 minutes/week, p < 0.05). The effects of commercial facility changes exhibit spatial heterogeneity: in low-accessibility areas, they promote both MVPA (42.67–70.02 minutes/week, p < 0.01) and total physical activity (68.35-88.00 minutes/week, p < 0.01), while showing diminishing effects in high-accessibility areas (from 33.79 minutes/week, p 0.05). The interaction between commercial disadvantage, public facilities, and time (DDD) reveals a positive impact on MVPA (71.20-89.96 minutes/week, p < 0.05), indicating that public sports facilities have stronger effects in commercially disadvantaged areas over time. This finding is further supported by our stratified analysis, which shows that public facilities have significant positive effects on MVPA (42.51–49.14 minutes/week, p < 0.001) in commercially disadvantaged areas, while showing no significant effects in commercially advantaged areas. These estimates remain robust after controlling for individual characteristics, built environment features, and park fixed effects. Discussion Main findings This study examined how different levels of neighborhood facility accessibility influenced residents' PA patterns during COVID-19 facility closures and reopening. Public facilities (parks and public sports facilities) demonstrated resilience with stable accessibility levels, while commercial facilities exhibited market-driven fluctuations, reflecting different operational mechanisms during the crisis. Facility changes affected PA recovery differently.(Larson et al., 2021 ). Our analysis revealed three key findings: First, our study confirms that higher facility accessibility is associated with increased PA duration. Specifically, in areas with high facility accessibility, participants' PA levels not only recovered but also exceeded pre-pandemic levels following the reopening of facilities. This finding aligns with previous research documenting PA recovery trends(Ding et al., 2020 ; Park et al., n.d.).For example, DID analysis revealed that areas with high park accessibility experienced greater increases in LPA duration ( β = 39.448, p = 0.032) and total PA ( β = 53.747, p = 0.020) compared to low-accessibility areas(Hoekman et al., 2023 ).Similarly, high public sports facility accessibility was associated with greater MVPA duration increases ( β = 39.069, p = 0.015) and total physical activity ( β = 48.791, p = 0.032). This functional differentiation supports theories on the multifaceted value of public facilities (Bergsgard et al., 2019 ) and confirms their fundamental role in promoting equitable PA recovery (Jones et al., 2009 ; Zhang et al., 2023 ). However, baseline commercial facility accessibility showed no significant association with PA changes (LPA: β =-8.843, p = 0.621; MVPA: β = 3.759, p = 0.814).This non-significant relationship between commercial facility accessibility and PA changes may be attributed to market saturation diminishing marginal returns, diverse exercise options in high-accessibility areas diluting impact, and facility quantity not reflecting quality or utilization patterns.(Müller et al., 2024 ) . Second, commercial facilities exhibited distinct market-oriented spatial patterns during the pandemic, with pronounced polarization effects: while high-accessibility areas maintained sufficient facility levels despite overall decreases, low commercial facility accessibility areas experienced further reductions in accessibility. (Shen et al., 2020 ). This spatial polarization manifested in differential impacts on MVPA recovery: residents in low-accessibility areas showed sensitivity to facility stable ( β = 70.019, p < 0.001), whereas those in high-accessibility areas demonstrated no significant association ( β = 19.708, p = 0.252) (Powell et al., 2006 ).This market-driven uneven distribution of commercial facilities not only exacerbated existing inequalities in PA recovery but also potentially limited facility investment in resource-deprived areas through the "Matthew Effect" (Rivera-Navarro et al., 2022 ; van Lenthe et al., 2005 ), consistent with previous findings on health resource spatial aggregation (Xu et al., 2022 ). Third, our study pioneered the quantification of public facilities' buffering effect through triple-difference analysis, revealing that public facilities can significantly buffer the negative impact of commercial facility disadvantages on MVPA duration ( β = 79.269, p = 0.050). This positive coefficient demonstrates that within the context of neighborhood deprivation, public facilities effectively mitigate the adverse effects of commercial facility disadvantages on physical activity patterns. Stratified analysis confirmed that public facilities positively promoted MVPA across all areas, but the effect was significant only in commercially disadvantaged regions ( β = 42.506, p = 0.001). This pattern reveals a conditional buffering effect: public facilities provide substantial protection against physical activity inequality in deprived neighborhoods while offering diminishing marginal returns in commercially advantaged environments. In areas with high public facility accessibility, residents maintained and even increased their MVPA levels (from 55.666 to 88.986 minutes/week) despite commercial facility disadvantages (Chastin et al., 2020 ) .(White et al., 2024 ). This buffering effect, particularly evident during market fluctuation, suggests that strategic public facility placement could serve as an effective intervention tool for reducing health inequities. (Milton et al., 2021 ; Rogers et al., 2024 ). Notably, areas with high public facility accessibility showed lower total PA duration despite supporting MVPA, consistent with previous findings(Yang et al., 2023 ). This counterintuitive result may be explained by residents' engagement in time-limited, structured activities rather than sustained LPA, particularly following the post-pandemic shift toward indoor exercise (Sun et al., 2024 ). Policy Implications and Practical Significance Based on our research findings, we propose three specific policy recommendations for improving urban PA facility provision: First, policymakers should continue integrating public sports facilities into basic public services, especially in areas with commercial facility disadvantages. Our study emphasizes the fundamental role of public PA facilities in promoting PA participation and their buffering effect against neighborhood health deprivation (Powell et al., 2006 ). This finding aligns with public health approaches emphasizing structural interventions to address health inequities rather than focusing solely on individual behavior change (Høyer-Kruse et al., 2024 ). Urban planning departments should: Incorporate PA facilities into basic livelihood security, with particular attention to facility allocation in deprived neighborhoods (Reijneveld et al., 2000 ; Rivera-Navarro et al., 2022 );For areas with limited fiscal resources, encourage the shared use of existing facilities such as school sports venues. This approach can provide affordable public facilities while optimizing resource allocation (Hoekman et al., 2023 ; Luo, 2018 ). Importantly, even in areas with high commercial facility density, public sports facilities still showed a positive effect, though not statistically significant. Deploying public sports facilities remains valuable in these areas, as low-income residents may be excluded from physical activity opportunities due to economic barriers to accessing commercial facilities(Higgerson et al., 2018 ). This strategic deployment ensures physical activity opportunities for all socioeconomic groups and prevents intra-neighborhood health inequalities(SFM et al., 2020 ). Second, urban planning departments should optimize the functional layout of PA facilities based on their differentiated effects. Our research reveals that different facility types influence specific dimensions of residents' PA behavior: parks and green spaces primarily promote LPA duration ( β = 39.448, p = 0.032) (Iamtrakul et al., 2024 ), while professional sports facilities are more conducive to increasing MVPA duration( β = 39.069, p = 0.015)(Jones et al., 2009 ). To meet residents' diverse PA needs(Eime et al., 2017 ; Zhang et al., 2023 ), we recommend incorporating professional exercise zones within parks and creating integrated facility networks that serve both casual and structured exercise purposes. This functional differentiation approach can maximize land use efficiency while addressing the full spectrum of PA needs. Third, local governments should develop collaborative management models that address the spatial polarization of PA resources. Our finding that commercial facilities show market-driven distribution patterns with diminishing returns in high-accessibility areas ( β = 19.708, p = 0.252) versus significant effects in low-accessibility areas ( β = 70.019, p < 0.001) suggests the need for targeted interventions: Implement conditional incentive policies (e.g., venue subsidies, tax benefits) specifically for commercial facility development in underserved areas, thereby improving facility accessibility and reducing spatial inequality(Humphreys & Zhou, 2015 ). Foster public-private partnerships through third-party organizations that can: Facilitate shared use agreements between sports clubs and public venues; Develop needs-based compensation mechanisms for facility providers in disadvantaged areas; Coordinate facility scheduling and maintenance to maximize utilization in areas with limited resources. This collaborative approach can enhance facility utilization while ensuring operational sustainability and spatial equity (Kang & Lee, 2022 ; Kenyon et al., 2018 ) Limitations Several limitations should be noted. First, our reliance on self-reported PA data may introduce recall bias, particularly given the retrospective nature of pandemic-related data collection. Second, despite using DID analysis to control for time-fixed effects, self-selection bias might exist where residents with stronger PA preferences choose neighborhoods with better facility access. Third, while our study in Shenzhen provides valuable insights, findings may not fully generalize to other urban contexts with different socioeconomic and built environment characteristics. Finally, our focus on facility accessibility does not capture other aspects such as facility quality and programming that may influence PA participation. Conclusion Through a natural experiment of facility closures during COVID-19, this study employed DID and DDD analyses to examine how different types of sports facilities influence PA duration. We found distinct patterns between public and commercial facilities. Commercial facilities showed market-driven spatial clustering with differential impacts on MVPA in high-accessibility ( β = 19.708, p = 0.252) versus low-accessibility areas ( β = 70.019, p < 0.001), suggesting diminishing marginal returns that potentially exacerbate health inequities. Public facilities demonstrated more balanced distribution and complementary effects: parks facilitated LPA duration ( β = 39.448, p = 0.032) and public sports facilities enhanced MVPA duration ( β = 39.069, p = 0.015). Through DDD analysis, we found that public sports facilities effectively buffered the negative impact of commercial facility disadvantages on MVPA duration ( β = 79.269, p = 0.050), with this buffering effect being particularly significant in neighborhoods experiencing commercial facility deprivation ( β = 42.506, p = 0.001). These findings provide evidence-based implications for addressing health inequities through environmental interventions and inform policies aimed at building resilient and equitable PA environments, particularly highlighting the value of strategic public facility investment in commercially disadvantaged neighborhoods.Our study identifies built environments as key social determinants of health, with public facilities' buffering effects supporting structural public health approaches to improve population-level physical activity and health outcomes. Declarations Ethics approval and consent to participate This study was approved by the Institutional Research Ethics Board at Harbin Institute of Technology. All participants were fully informed about the design and the purpose of the study, and they provided informed consent before participating. The research was conducted in accordance with the principles of the Declaration of Helsinki. Consent for publication Not applicable. Availability of data and materials Due to privacy and confidentiality agreements, the raw dataset from this study is not publicly available. However, aggregated data can be made available from the corresponding author upon reasonable request. Competing interests The authors declare that they have no competing interests. Funding This work was supported by the National Natural Science Foundation of China [grant number 42371238], the Guangdong Province Philosophy and Social Sciences Planning Project [grant number GD23XGL087], and the Shenzhen Science and Technology Innovation Commission [grant number 0231129125223001]. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Authors' contributions PZ conceptualized the study idea, provided methodological guidance, data resources, and critically reviewed and revised the manuscript. JY performed data analysis and wrote the original draft of the manuscript. All authors read and approved the final manuscript. Acknowledgements We would like to express our incere thanks to Yirou Chen, Xinsu Lv, and ZhenHu for their dedication and assistance during the fieldwork and data collection. This work was supported by the National Natural Science Foundation of China under Grant [number 42371238], the Guangdong Province Philosophy and Social Sciences Planning Project [number GD23XGL087], and the Shenzhen Science and Technology Innovation Commission under Grant [number 0231129125223001]. References Ainsworth BE. (2024). 2024 Adult Compendium – Compendium of Physical Activities . https://pacompendium.com/adult-compendium/ Bergsgard NA, Borodulin K, Fahlen J, Hoyer-Kruse J, Iversen EB. National structures for building and managing sport facilities: A comparative analysis of the Nordic countries. SPORT Soc. 2019;22(4):525–39. https://doi.org/10.1080/17430437.2017.1389023 . Chastin SFM, Van Cauwenberg J, Maenhout L, Cardon G, Lambert E, Van Dyck D. Inequality in physical activity, global trends by income inequality and gender in adults. Int J Behav Nutr Phys ACTIVITY. 2020;17(1):142. https://doi.org/10.1186/s12966-020-01039-x . Culture R. Television, Tourism andSports Bureau of Shenzhen Municipality. (2019). List of Parks in Shenzhen . https://cgj.sz.gov.cn/gkmlpt/content/5/5402/post_5402966.html#476 Ding D, del Pozo Cruz B, Green MA, Bauman AE. Is the COVID-19 lockdown nudging people to be more active: A big data analysis. Br J Sports Med. 2020;54(20):1183–1183. https://doi.org/10.1136/bjsports-2020-102575 . Durão S, Burns J, Schmidt B-M, Tumusiime D, Hohlfeld A, Pfadenhauer L, Ongolo-Zogo C, Rehfuess E, Kredo T. Infrastructure, policy and regulatory interventions to increase physical activity to prevent cardiovascular diseases and diabetes: A systematic review. BMC Public Health. 2023;23:112. https://doi.org/10.1186/s12889-022-14841-y . Eime RM, Harvey J, Charity MJ, Casey M, Westerbeek H, Payne WR. The relationship of sport participation to provision of sports facilities and socioeconomic status: A geographical analysis. Aust N Z J Public Health. 2017;41(3):248–55. https://doi.org/10.1111/1753-6405.12647 . Eriksson U, Arvidsson D, Sundquist K. Availability of exercise facilities and physical activity in 2,037 adults: Cross-sectional results from the Swedish neighborhood and physical activity (SNAP) study. BMC Public Health. 2012;12:607. https://doi.org/10.1186/1471-2458-12-607 . Esri. (2020). ArcGIS Desktop: Release 10.8. Redlands, CA: Environmental Systems Research Institute[^121^]. Esri . https://www.esri.com/about/newsroom/announcements/esri-releases-guide-teaching-arcgis-desktop-10-8/ Gidlow C, Cerin E, Sugiyama T, Adams MA, Mitas J, Akram M, Reis RS, Davey R, Troelsen J, Schofield G, Sallis JF. Objectively measured access to recreational destinations and leisure-time physical activity: Associations and demographic moderators in a six-country study. Health Place. 2019;59:102196. https://doi.org/10.1016/j.healthplace.2019.102196 . Gomes TN, Suikkanen S, Thuany M, Vaananen I, Donnelly A. Physical activity in outdoor spaces, sleep quality and general health: A pilot feasibility study. Lancet (London England). 2023;402(Suppl 1):S45. https://doi.org/10.1016/S0140-6736(23)02094-9 . Higgerson J, Halliday E, Ortiz-Nunez A, Brown R, Barr B. Impact of free access to leisure facilities and community outreach on inequalities in physical activity: A quasi-experimental study. J Epidemiol Commun Health. 2018;72(3):252–8. https://doi.org/10.1136/jech-2017-209882 . Hoekman R, Grubben M, Kraaykamp G. (2023). Sport club dropout under COVID-19 in the Netherlands: Do characteristics of the neighbourhood matter? Frontiers in Sports and Active Living , 5 . https://doi.org/10.3389/fspor.2023.1168608 Høyer-Kruse J, Schmidt EB, Hansen AF, Pedersen MRL. The interplay between social environment and opportunities for physical activity within the built environment: A scoping review. BMC Public Health. 2024;24(1):2361. https://doi.org/10.1186/s12889-024-19733-x . Humphreys BR, Zhou L. Sports facilities, agglomeration, and public subsidies. Reg Sci Urban Econ. 2015;54:60–73. https://doi.org/10.1016/j.regsciurbeco.2015.07.004 . Iamtrakul P, Chayphong S, Gao W. Assessing spatial disparities and urban facility accessibility in promoting health and well-being. Transp Res Interdisciplinary Perspect. 2024;25:101126. https://doi.org/10.1016/j.trip.2024.101126 . Jiang Y, Liu Y, Liu Z, Wang C, Shi Z, Zhao H, Sun D, Sun W, Wang X. Spatial Distribution Characteristics of Public Fitness Venues: An Urban Accessibility Perspective. SUSTAINABILITY. 2023;15(1):601. https://doi.org/10.3390/su15010601 . Jing Y, Liu Y, Cai E, Liu Y, Zhang Y. Quantifying the spatiality of urban leisure venues in Wuhan, Central China—GIS-based spatial pattern metrics. SUSTAINABLE CITIES Soc. 2018;40:638–47. https://doi.org/10.1016/j.scs.2018.05.015 . Jones A, Hillsdon M, Coombes E. Greenspace access, use, and physical activity: Understanding the effects of area deprivation. Prev Med. 2009;49(6):500–5. https://doi.org/10.1016/j.ypmed.2009.10.012 . Jowshan M-R, Pourjavid A, Amirkhizi F, Hosseini M-H, Zolghadrpour M-A, Hamedi-Shahraki S, Motlagh AH, Asghari S. Adherence to combined healthy lifestyle and odds of metabolic syndrome in Iranian adults: The PERSIAN Dena cohort study. Sci Rep. 2025;15(1):5164. https://doi.org/10.1038/s41598-025-89028-7 . Kahneman D, Krueger AB, Schkade DA, Schwarz N, Stone AA. A Survey Method for Characterizing Daily Life Experience: The Day Reconstruction Method. Science. 2004;306(5702):1776–80. https://doi.org/10.1126/science.1103572 . Kang M, Lee Y. The Gap in Community Sports: Utilization of Sports Facilities in South Korea. Int J Environ Res Public Health. 2022;19(8). https://doi.org/10.3390/ijerph19084495 . Article 8. Kegler MC, Gauthreaux N, Hermstad A, Arriola KJ, Mickens A, Ditzel K, Hernandez C, Haardoerfer R. Inequities in Physical Activity Environments and Leisure-Time Physical Activity in Rural Communities. Prev Chronic Dis. 2022;19:210417. https://doi.org/10.5888/pcd19.210417 . Kenyon JA, Mason C, Rookwood J. Emerging third-sector sports organisations and navigating uncertainty in an era of austerity: A single ethnographic case study from Liverpool. Int J SPORT POLICY POLITICS. 2018;10(1):25–42. https://doi.org/10.1080/19406940.2018.1425732 . Kohl HW, Craig CL, Lambert EV, Inoue S, Alkandari JR, Leetongin G, Kahlmeier S. The pandemic of physical inactivity: Global action for public health. Lancet. 2012;380(9838):294–305. https://doi.org/10.1016/S0140-6736(12)60898-8 . Larson LR, Zhang Z, Oh JI, Beam W, Ogletree S, Bocarro J, Lee K, Casper JM, Stevenson K, Hipp J, Mullenbach LE, Carusona M, Wells ML. Urban Park Use During the COVID-19 Pandemic: Are Socially Vulnerable Communities Disproportionately Impacted? Front Sustainable Cities. 2021. https://doi.org/10.3389/frsc.2021.710243 . Lee S, Lee C, Xu M, Li W, Ory M. People living in disadvantaged areas faced greater challenges in staying active and using recreational facilities during the COVID-19 pandemic. Health Place. 2022;75:102805. https://doi.org/10.1016/j.healthplace.2022.102805 . Li B, Valerino-Perea S, Zhou W, Xie Y, Syrett K, Peters R, He Z, Zou Y, de Vocht F, Foster C. The impact of the world’s first regulatory, multi-setting intervention on sedentary behaviour among children and adolescents (ENERGISE): A natural experiment evaluation. Int J Behav Nutr Phys ACTIVITY. 2024;21(1):53. https://doi.org/10.1186/s12966-024-01591-w . Lisbona Guillén M, Sánchez Martín J, Sánchez Martín R. Urban segregation and body differentiation: Private gyms in Merida (Mexico). Iztapalapa Revista de ciencias sociales y humanidades. 2024;45(96):235–63. https://doi.org/10.28928/ri/962024/aot6/lisbona/etal/x . Liu M, Timmermans EJ, Zou D, Grobbee DE, Zhou S, Vaartjes I. Impact of green space exposure on blood pressure in Guangzhou, China: Mediation by air pollution, mental health, physical activity, and weight status. Environ Pollut. 2024;356:124251. https://doi.org/10.1016/j.envpol.2024.124251 . Liu Y, Wang X, Zhou S, Wu W. The association between spatial access to physical activity facilities within home and workplace neighborhoods and time spent on physical activities: Evidence from Guangzhou, China. Int J Health Geogr. 2020;19(1):22. https://doi.org/10.1186/s12942-020-00216-2 . Lundh H, Arvidsson D, Greven C, Fridolfsson J, Börjesson M, Boman C, Lauruschkus K, Lundqvist S, Melin K, Bernhardsson S. Physical activity and sedentary behaviour amongst children with obesity—Exploring cross-sectional associations between child and parent. J Activity Sedentary Sleep Behav. 2025;4(1):2. https://doi.org/10.1186/s44167-025-00072-0 . Luo Q. (2018). Research on the Predicament and Optimization of the Openness of College Sports Facilities in China. In Y. M. Wang & D. C. Zhou, editors, 2018 4TH INTERNATIONAL CONFERENCE ON EDUCATION, MANAGEMENT AND INFORMATION TECHNOLOGY (ICEMIT 2018) (pp. 992–996). Francis Acad Press. https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000450415100224 Macdonald L. Associations between spatial access to physical activity facilities and frequency of physical activity; how do home and workplace neighbourhoods in West Central Scotland compare? Int J Health Geogr. 2019;18:2. https://doi.org/10.1186/s12942-019-0166-z . Menardo E, De Dominicis S, Pasini M. Exploring Perceived and Objective Measures of the Neighborhood Environment and Associations with Physical Activity among Adults: A Review and a Meta-Analytic Structural Equation Model. Int J Environ Res Public Health. 2022;19(5):2575. https://doi.org/10.3390/ijerph19052575 . Milton K, Cavill N, Chalkley A, Foster C, Gomersall S, Hagstromer M, Kelly P, Kolbe-Alexander T, Mair J, McLaughlin M, Nobles J, Reece L, Shilton T, Smith BJ, Schipperijn J. Eight Investments That Work for Physical Activity. J Phys Act Health. 2021;18(6):625–30. https://doi.org/10.1123/jpah.2021-0112 . Müller C, Paulsen L, Bucksch J, Wallmann-Sperlich B. Built and natural environment correlates of physical activity of adults living in rural areas: A systematic review. Int J Behav Nutr Phys Activity. 2024;21(1):52. https://doi.org/10.1186/s12966-024-01598-3 . O’Brien D. Commercial sport organisations and community capacity building: A case study of two surf parks. SPORT Manage Rev. 2021;24(5):723–46. https://doi.org/10.1080/14413523.2021.1902143 . Olden A, Møen J. The triple difference estimator. Econometrics J. 2022;25(3):531–53. https://doi.org/10.1093/ectj/utac010 . Olvera Alvarez HA, Appleton AA, Fuller CH, Belcourt A, Kubzansky LD. An Integrated Socio-Environmental Model of Health and Well-Being: A Conceptual Framework Exploring the Joint Contribution of Environmental and Social Exposures to Health and Disease Over the Life Span. Curr Environ Health Rep. 2018;5(2):233–43. https://doi.org/10.1007/s40572-018-0191-2 . Park AH, Zhong S, Yang H, Jeong J, Lee C (n.d.), editors. Impact of COVID-19 on physical activity: A rapid review. Journal of Global Health , 12 , 05003. https://doi.org/10.7189/jogh.12.05003 Powell LM, Slater S, Chaloupka FJ, Harper D. Availability of Physical Activity–Related Facilities and Neighborhood Demographic and Socioeconomic Characteristics: A National Study. Am J Public Health. 2006;96(9):1676–80. https://doi.org/10.2105/AJPH.2005.065573 . Reijneveld SA, Verheij RA, de Bakker DH. The impact of area deprivation on differences in health: Does the choice of the geographical classification matter? J Epidemiol Commun Health. 2000;54(4):306–13. https://doi.org/10.1136/jech.54.4.306 . Rivera-Navarro J, Bonilla L, Gullon P, Gonzalez-Salgado I, Franco M. Can we improve our neighbourhoods to be more physically active? Residents’ perceptions from a qualitative urban health inequalities study. Health Place. 2022;77:102658. https://doi.org/10.1016/j.healthplace.2021.102658 . Rocha V, Ribeiro AI, Severo M, Barros H, Fraga S. Neighbourhood socioeconomic deprivation and health-related quality of life: A multilevel analysis. PLoS ONE. 2017;12(12):e0188736. https://doi.org/10.1371/journal.pone.0188736 . Rogers AE, Schenkelberg MA, Stoepker P, Westmark D, Srivastava D, Dzewaltowski DA. Indicators of community physical activity resources and opportunities and variation by community sociodemographic characteristics: A scoping review. Prev Med Rep. 2024;40:102656. https://doi.org/10.1016/j.pmedr.2024.102656 . Salvo G, Lashewicz BM, Doyle-Baker PK, McCormack GR. Neighbourhood Built Environment Influences on Physical Activity among Adults: A Systematized Review of Qualitative Evidence. Int J Environ Res Public Health. 2018;15(5):897. https://doi.org/10.3390/ijerph15050897 . Santacruz Lozano JA, Espada Mateos M, Clemente Remon A, Navarro J-B. J. E. (2021). Spanish sport facilities: Differences between public and private, and according to their business model. RETOS-NUEVAS TENDENCIAS EN EDUCACION FISICA DEPORTE Y RECREACION , 39 , 38–45. Sawyer A, Ucci M, Jones R, Smith L, Fisher A. Simultaneous evaluation of physical and social environmental correlates of physical activity in adults: A systematic review. SSM-POPULATION HEALTH. 2017;3:506–15. https://doi.org/10.1016/j.ssmph.2017.05.008 . SFM C, Van Cauwenberg J, Maenhout L, Cardon G, Lambert EV, Van Dyck D. Inequality in physical activity, global trends by income inequality and gender in adults. Int J Behav Nutr Phys Activity. 2020;17(1):142. https://doi.org/10.1186/s12966-020-01039-x . Shen J, Cheng J, Huang W, Zeng F. An Exploration of Spatial and Social Inequalities of Urban Sports Facilities in Nanning City. China Sustain. 2020;12(11):4353. https://doi.org/10.3390/su12114353 . Singer JD, Willett JB. (2003). Applied longitudinal data analysis: Modeling change and event occurrence (pp. xx, 644). Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195152968.001.0001 StataCorp. (2021). Stata Statistical Software: Release 17.College Station, TX: StataCorp LLC. https://stata-nz.com/docs/manuals/stata17/p.pdf Stokols D. (1996). Translating social ecological theory into guidelines for community health promotion. AMERICAN JOURNAL OF HEALTH PROMOTION , 10 (4), 282–298. Art and Science of Health Promotion Conference. https://doi.org/10.4278/0890-1171-10.4.282 Sun P, Sun J, Jin L, Zhu Y. COVID-19 pandemic changes the outdoor physical activity preference in Chinese city: A 7-year GPS trajectory data analysis. Cities. 2024;152:105253. https://doi.org/10.1016/j.cities.2024.105253 . van Lenthe FJ, Brug J, Mackenbach JP. Neighbourhood inequalities in physical inactivity: The role of neighbourhood attractiveness, proximity to local facilities and safety in the Netherlands. Soc Sci Med. 2005;60(4):763–75. https://doi.org/10.1016/j.socscimed.2004.06.013 . Wang Y, Xie H, Hua Y, Jin B, Qiu Z. Theoretical framework of life circles in Chinese small towns and the optimization of spatial layout for public service facilities based on residents’ distance sensitivity. Humanit Social Sci Commun. 2024;11(1):1–14. https://doi.org/10.1057/s41599-024-04321-6 . White RL, Vella S, Biddle S, Sutcliffe J, Guagliano JM, Uddin R, Burgin A, Apostolopoulos M, Nguyen T, Young C, Taylor N, Lilley S, Teychenne M. Physical activity and mental health: A systematic review and best-evidence synthesis of mediation and moderation studies. Int J Behav Nutr Phys ACTIVITY. 2024;21(1):134. https://doi.org/10.1186/s12966-024-01676-6 . WHO. (2020). WHO guidelines on physical activity and sedentary behaviour . https://www.who.int/publications/i/item/9789240015128 Xu R, Yue W, Wei F, Yang G, Chen Y, Pan K. Inequality of public facilities between urban and rural areas and its driving factors in ten cities of China. Sci Rep. 2022;12(1):13244. https://doi.org/10.1038/s41598-022-17569-2 . Yang, Liu. An age-stratified study on the accessibility to sports venues and their impact on resident satisfaction: Evidence from urban Beijing. Progress Geogr. 2024;43(7):1416–28. https://doi.org/10.18306/dlkxjz.2024.07.011 . Yang Y, Peng C, Yeung CY, Ren C, Luo H, Lu Y, Yip PSF, Webster C. Moderation effect of visible urban greenery on the association between neighbourhood deprivation and subjective well-being: Evidence from Hong Kong. Landsc Urban Plann. 2023;231:104660. https://doi.org/10.1016/j.landurbplan.2022.104660 . Zhang Y, Koene M, Chen C, Wagenaar C, Reijneveld SA. Associations between the built environment and physical activity in children, adults and older people: A narrative review of reviews. Prev Med. 2024;180:107856. https://doi.org/10.1016/j.ypmed.2024.107856 . Zhang Y, Ming Y, Y., Shi B. Spatial distribution characteristics and causes of public sports venues in China. Sci Rep. 2023;13(1):15056. https://doi.org/10.1038/s41598-023-42308-6 . Zhou P, Hu Z, Chen Y, Liu K, Wang Y. Parenthood, spatial temporal environmental exposure, and leisure-time physical activity participation: Evidence from a micro-timescale retrospective longitudinal study. Health Place. 2024;85:103170. https://doi.org/10.1016/j.healthplace.2023.103170 . Additional Declarations No competing interests reported. Supplementary Files AdditionalFile1Supplementmaterial.doc AdditionalFile2STROBEchecklistv4cohort.doc Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6484311","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":457312622,"identity":"55efd61d-0c9b-40fa-aa60-f2ea57df4399","order_by":0,"name":"Jingyi Yang","email":"","orcid":"","institution":"Harbin Institute of Technology Shenzhen","correspondingAuthor":false,"prefix":"","firstName":"Jingyi","middleName":"","lastName":"Yang","suffix":""},{"id":457312623,"identity":"652b8657-1df9-4287-aa13-a42bcf459600","order_by":1,"name":"Peiling Zhou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA70lEQVRIiWNgGAWjYDACCQgpw8bAwPiYwQDESSBOCw9QC7Mxg4EB0VoYeICYTZqBgQgt8rObj0kXVFjw8Em3X6suKPjDwM+eY8DwcwduLYxzjqVJzzgDdJjMmbLbM4AOk+x5Y8DYewa3FmaJHDNp3jagFomctNs8QC0GN3IMmBnbcGthA2v5B9FSDNJiT0gLD1hLA0hL+jFmsC0SBLRISKQlW884BraFWZrHwJhH4syzgoO9eLTIz0g+eLugpk5Ofkb6w888f+Tk+NuTNz74iUcLOAigbgRHPSiCGA7g1wDXwv6AkMJRMApGwSgYoQAANYY+OQePmuwAAAAASUVORK5CYII=","orcid":"","institution":"Harbin Institute of Technology Shenzhen","correspondingAuthor":true,"prefix":"","firstName":"Peiling","middleName":"","lastName":"Zhou","suffix":""}],"badges":[],"createdAt":"2025-04-19 11:08:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6484311/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6484311/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":83143771,"identity":"f8ebdc64-be05-496e-bd3f-06848fdcbaac","added_by":"auto","created_at":"2025-05-20 12:38:55","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":360936,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eResearch Design: COVID-19 as a natural experiment to evaluate facility accessibility impacts on physical activity and health equity\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eArrows represent causal relationships between different variables. Green arrows indicate statistically significant positive relationships, red arrows indicate statistically significant negative relationships, gray dashed arrows indicate non-significant relationships, and orange arrows represent buffering effects. In our analysis, the buffering effect refers to how the presence of public sports facilities mitigated the negative impact of commercial facility reduction on moderate-to-vigorous physical activity in areas with originally low commercial facility accessibility that experienced further decreases.\u003c/p\u003e","description":"","filename":"F1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6484311/v1/04c5aa46807f6fa973d02cd4.jpeg"},{"id":83143753,"identity":"714d579b-3e7c-44bb-8403-f7473bec2c3b","added_by":"auto","created_at":"2025-05-20 12:38:50","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":452804,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSpatial Distribution of Sports Facilities in Shenzhen: Visualizing inequities in community-level physical activity resources\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSpatial distribution maps showing density patterns of physical activity facilities across Shenzhen's community committee boundaries during 2019-2020. Panels a-b display commercial sports facilities in 2019 and 2020 respectively, using orange-red graduated colors (0-12, 13-43, 44-78, 79-122, 123-199 facilities/km²). Panels c-d show public sports facilities in 2019 and 2020 respectively, with purple graduated colors (1-2, 3-4, 5-8, 9-17 facilities/km²). Panels e-f illustrate park distribution in 2019 and 2020 respectively, using teal-green graduated colors (0-1, 2-4, 5-14, 15-30 facilities/km²). For all panels, darker shades indicate higher facility density. These maps demonstrate the spatial clustering patterns of different facility types and their changes between pre-pandemic and pandemic periods.\u003c/p\u003e","description":"","filename":"F2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6484311/v1/f359927d414945be00e171d8.jpeg"},{"id":104426214,"identity":"6ccf613a-df9f-4436-8d3b-b8315567aa04","added_by":"auto","created_at":"2026-03-11 14:42:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2388659,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6484311/v1/55856906-afec-4f93-a5f8-c7e564ba13bf.pdf"},{"id":83143754,"identity":"a4662ff8-48a4-46b2-9977-1fca365ad81e","added_by":"auto","created_at":"2025-05-20 12:38:51","extension":"doc","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":207872,"visible":true,"origin":"","legend":"","description":"","filename":"AdditionalFile1Supplementmaterial.doc","url":"https://assets-eu.researchsquare.com/files/rs-6484311/v1/76374e5f50bed9b75a505916.doc"},{"id":83143764,"identity":"371988ea-5b5a-4611-9a50-1743a9bc18df","added_by":"auto","created_at":"2025-05-20 12:38:53","extension":"doc","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":84313,"visible":true,"origin":"","legend":"","description":"","filename":"AdditionalFile2STROBEchecklistv4cohort.doc","url":"https://assets-eu.researchsquare.com/files/rs-6484311/v1/9754c238c4f36b020d80b990.doc"}],"financialInterests":"No competing interests reported.","formattedTitle":"Buffering effects of public sports facilities on physical activity and health equity during commercial facility fluctuations: A natural experiment in Shenzhen, China","fulltext":[{"header":"Introduction","content":"\u003cp\u003eExtensive research has demonstrated the negative impact of physical inactivity on non-communicable diseases, emphasizing the importance of boosting physical activity levels across populations to enhance public health(Kohl et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). According to the socioecological model, environmental factors play a crucial role affecting PA behaviors(Stokols, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). Facilities at neighborhood serve as primary venues for residents' PA engagement, and their accessibility is strongly associated with population health status(Salvo et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Comprehensive PA infrastructure, such as sports venues, fitness rooms, and public green Spaces, creates opportunities for PA participation(Dur\u0026atilde;o et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), thereby enhancing physical fitness, preventing non-communicable diseases, and improving mental health outcomes(Gomes et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Menardo et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; White et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Neighborhood deprivation, including PA facility scarcity in some neighborhood contributes to health inequities(Kegler et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Rocha et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Sawyer et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Residents in communities with low accessibility to PA facilites exhibit lower PA levels and ambulatory activities compared to neighborhoods with high sports facility accessibility(Eriksson et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; van Lenthe et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), and thus have poorer health outcomes(Lee et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Reijneveld et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2000\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMore recent models emphasize the combined impacts between environmental and social factors upon health outcomes(Olvera Alvarez et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). For example, public and commercial PA facilities in neoliberal cities shows divergent patterns of spatial distribution and serve for different groups of populations (O\u0026rsquo;Brien, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Santacruz Lozano et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Public PA facilities, most of which are government-subsidized and accessible to the general public, are designed to offer PA opportunities for a wider population. According to WHO Guidelines on Physical Activity and Sedentary Behavior(2020), governments in many countries employ these facilities as interventions to promote PA duration, although their availability is typically limited by resource allocation constraints (Bergsgard et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Kang \u0026amp; Lee, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In contrast, commercial facilities may cause inequal and even bipolar health outcomes between populations of different socioeconomic status. Funded and operated by private entities and accessible only to members, commercial facilities tend to offer specialized and diverse sports programs. These facilities primarily cater to individuals willing to pay for particular services (Kang \u0026amp; Lee, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Rivera-Navarro et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) and are often concentrated in areas of higher socioeconomic status due to market dynamics (Lisbona Guill\u0026eacute;n et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn China, while the central and local governments have taken multiple measures to construct PA facilities in urban neighborhoods to intervene physical inactivity and to promote public health, limited financial expenditure on health intervention and the marketization of public health industry cause spatial inequities in public and commercial PA facility accessibility at different geographic scales, across regions(Jiang et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), between urban and rural areas (Jing et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), across and within cities (Xu et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) and districts(Yang \u0026amp; Liu, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). These multidimensional aspects of neighborhood deprivation contribute to health inequities through disparate access to PA opportunities(Jiang et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Y. Liu et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) .\u003c/p\u003e \u003cp\u003eWhile the uneven distribution of public and commercial PA facilities are simultaneously causing health inequalities, little research has simultaneously investigated the integrated effect of both public and commercial PA facilities(Macdonald, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) nor fully elucidated the causal effects of varying sports facility accessibility on PA behaviors (Sawyer et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The closure and subsequent reopening of sports facilities during the COVID-19 pandemic (Hoekman et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Lee et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) present an ideal natural experimental setting to investigate the relationship between sports facility accessibility and PA levels (Ding et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This study examines the neighborhood-level PA facilities and their impacts on residential PA patterns to provide evidence-based recommendations for equitable sports facility planning in urban environments. In particular, the objectives and hypotheses of this study include:\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eObjective 1\u003c/strong\u003e \u003cp\u003eTo examine how inequitable distribution of neighborhood sports facilities affects residents' PA duration disparities\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eHypothesis 1\u003c/strong\u003e \u003cp\u003eThe spatial inequality in sports facility distribution creates persistent differences in neighborhood PA duration. Areas with higher facility accessibility demonstrate greater PA duration and faster recovery following facility reopening, with parks primarily facilitating LPA and sports venues promoting MVPA.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eObjective 2\u003c/strong\u003e \u003cp\u003eTo investigate how changes in commercial sports facility accessibility polarize residents' PA duration\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eHypothesis 2\u003c/strong\u003e \u003cp\u003eChanges in commercial facility accessibility create variations in MVPA duration. This effect is particularly pronounced in areas with limited commercial facility accessibility but negligible in areas with adequate commercial sports resources.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eObjective 3\u003c/strong\u003e \u003cp\u003eTo evaluate how public facilities buffer against PA duration inequities caused by spatial clustering of commercial facilities\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eHypothesis 3\u003c/strong\u003e \u003cp\u003eIn areas with limited commercial facility accessibility, public sports facilities mitigate the impact of commercial facility fluctuations on residents' PA duration by providing alternative venues particularly in communities experiencing reductions or scarcity in commercial facilities.\u003c/p\u003e \u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eResearch Design\u003c/h2\u003e \u003cp\u003eThis study employed a natural experimental design, utilizing operational changes in commercial sports facilities during the COVID-19 pandemic as a quasi-natural intervention. We conducted a retrospective longitudinal assessment in November 2020 (T2), collecting both current data and recalled baseline data from November 2019 (T1). A difference-in-differences (DID) approach was applied to evaluate the impact of changes in facility accessibility on PA behavior, allowing us to identify intervention effects while controlling for temporal trends and unobserved confounders. The temporal framework encompassed: (1) baseline period (T1) in November 2019 when facilities operated normally; (2) facility closure phase beginning in late January 2020 due to COVID-19 containment measures; (3) differential reopening phase from March 2020, with public facilities gradually resuming operations while some commercial facilities permanently closed; and (4) Stable new facility landscape with full restoration of public facilities in November 2020 (T2). The matching seasons (November) in baseline and follow-up measurements helped control for seasonal variation in PA duration. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e illustrates our natural experiment framework, showing the baseline and follow-up conditions, the intervention point created by the pandemic, and assess how changes in facility accessibility affected physical activity patterns.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eStudy area and sampling procedures\u003c/h3\u003e\n\u003cp\u003eA multi-stage sampling strategy was employed to recruit participants from 23 residential communities across six administrative districts in Shenzhen, China, ensuring geographic and socioeconomic representation. Participants lived in a variety of housing types including commercial housing, urban villages, danwei compounds, and public housing. we recruited a minimum of 30 eligible participants aged 18\u0026ndash;65 years from each community. A total of 701 participants provided valid responses in November 2020, including their current PA duration per week and those retrospectively reported in November 2019. The sample composition reflected the distribution of sex, age, and occupation. All participants provided written informed consent, and the study protocol was approved by the Ethics Committee of the University that the first author is studying at.\u003c/p\u003e\n\u003ch3\u003eData Collection\u003c/h3\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003ePhysical Activity Data Collection Protocol\u003c/h2\u003e \u003cp\u003ePA data were collected using the Day Reconstruction Method (DRM), following protocols detailed in our previous work(Kahneman et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Zhou et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).Trained interviewers conducted standardized face-to-face interviews between November and December 2020, where participants recalled their typical weekday activity patterns for November 2019 (T1) and November 2020 (T2). To minimize recall bias, participants verified their activity information against smartphone payment records and mobility trajectories.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eEnvironmental Facility Data\u003c/h3\u003e\n\u003cp\u003eEnvironmental Facility Data were collected from multiple authoritative sources. Public facilities records, including parks and sports venues, were obtained from the website of Culture, Radio, Television, Tourism and Sports Bureau of Shenzhen Municipality and Urban Management And Law Enforcement (2019). Commercial sports facility records were extracted from Gaode Maps Points of Interest (POI) database at two time points: baseline (T1: November 2019) and follow-up (T2: November 2020) and were triangulated with facility utilization records from mobile phone data. All facility locations were geocoded using a standardized protocol to calculate spatial accessibility indices.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eMeasurements\u003c/h2\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003ePhysical Activity Classification and Assessment\u003c/h2\u003e \u003cp\u003eWeekly PA duration served as the primary outcome measure. Activities were categorized based on metabolic equivalents (METs) into moderate-to-vigorous physical activity(MVPA)(METs\u0026thinsp;\u0026ge;\u0026thinsp;3.0, such as gym workouts, ball sports, and running) and light physical activity(LPA) (1.5\u0026thinsp;\u0026lt;\u0026thinsp;METs\u0026thinsp;\u0026lt;\u0026thinsp;3.0, such as walking and Walk the dog)(Ainsworth, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Participants reported activity type, location, weekly frequency, and duration per session across three daily periods (morning, noon, evening). Weekly PA duration per week was calculated by multiplying frequency with per-session duration. From 921 initially collected activity records, 841 valid records from 701 participants were retained after excluding samples missing critical information, or falling without the neighborhood scope.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e\n\u003ch3\u003eAccessibility of Sports Facilities\u003c/h3\u003e\n\u003cp\u003eFollowing Gidlow (2019)and M. Liu et al (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), the accessibility of PA facilities was measured by the number of facilities within 1,000-meter residential catchment areas with China's urban planning policy for \"15-minute community life circles\". Environmental intervention indicators were defined based on facility density changes during COVID-19. We measured the number of three types of facilities within the 1000m residential buffer, and defined the high-low level of facilities according to the median values of each density. Two types of public facilities were considered: (1) park accessibility: high-density areas defined as \u0026ge;\u0026thinsp;4 facilities; (2) public sports facility accessibility: categorized by presence (\u0026ge;\u0026thinsp;1 facility).For commercial sports facilities, which served as the primary intervention variable, baseline high-density areas were defined as \u0026gt;\u0026thinsp;43 facilities, with subsequent change status classified as stable (reduction\u0026thinsp;\u0026le;\u0026thinsp;10% from baseline) or decline (reduction\u0026thinsp;\u0026gt;\u0026thinsp;10%), The 10% threshold was established through empirical analysis as a natural break in facility change distribution, effectively distinguishing between normal market fluctuations (\u0026lt;\u0026thinsp;10%) and substantial changes (\u0026gt;\u0026thinsp;10%) that significantly impact resident behaviors(Park et al., n.d.; Zhang et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2024\u003c/span\u003e)\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eCovariates\u003c/h2\u003e \u003cp\u003eOur analysis incorporated two sets of covariates: individual characteristics and built environmental features. Individual variables included individual characteristic: age (categorized as \u0026lt;\u0026thinsp;18, 18\u0026ndash;44, 45\u0026ndash;64, \u0026gt;\u0026thinsp;65 years), gender (male\u0026thinsp;=\u0026thinsp;0, female\u0026thinsp;=\u0026thinsp;1), educational attainment (high school or below =\u0026thinsp;0, bachelor's degree or above =\u0026thinsp;1), and annual household income. These categorizations align with previous studies examining associations between built environment and PA behaviors (Jowshan et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Lundh et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Built environmental-level characteristics were measured within participants' walking catchment areas using ArcGIS 10.8 (Esri, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). These measures included street connectivity, building density, and land-use mix.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eWe employed quasi-experimental methods - Difference-in-Differences (DID) (Singer \u0026amp; Willett, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2003\u003c/span\u003e)and Triple Differences (DDD)(Olden \u0026amp; M\u0026oslash;en, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) approaches to evaluate how different levels of facility accessibility causally affect PA duration. These methods allow us to estimate the impact of facility accessibility changes by controlling for time-invariant unobserved heterogeneity and time-specific shocks.\u003c/p\u003e \u003cp\u003eIn all models:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e\u0026bull; \u003cem\u003eYit\u003c/em\u003e represents individual i's weekly PA duration at \u003cem\u003eTime\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e, analyzed separately for:\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003ea) Light Physical Activity (LPA) Duration\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eb) Moderate-to-Vigorous Physical Activity (MVPA) Duration\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003ec) Total Physical Activity Duration\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e\u0026bull; \u003cem\u003eTime\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e indicates pre (T1) and post (T2) periods\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e\u0026bull; \u003cem\u003eγXit\u003c/em\u003e controls for baseline effects(time, pub, com), built environment features (population density, land-use mix, street connectivity), and individual characteristics (age, gender, education, income)\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e\u0026bull; \u003cem\u003eαpark_j\u003c/em\u003e accounts for park-specific fixed effects\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e\u0026bull; \u003cem\u003eεit\u003c/em\u003e represents the error term\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eParallel Trend Assumption\u003c/h2\u003e \u003cp\u003eWe assume that, in the absence of any facility-related interventions (including changes in facility numbers or operational status), the PA duration would follow similar trends across different areas over time. That is, if facility accessibility had no causal impact on PA duration, the change in PA duration should not differ across these areas before and after the intervention. The net effect was calculated as:\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eimpact_of_facility = (PA\u003csub\u003ehigh_access,t2\u003c/sub\u003e - PA\u003csub\u003ehigh_access,t1\u003c/sub\u003e) - (PA\u003csub\u003elow_access,t2\u003c/sub\u003e - PA\u003csub\u003elow_access,t1\u003c/sub\u003e)\u003c/h2\u003e \u003cp\u003ewhere PA is the outcome of physical activity in the control group (low accessibility areas) vs. the treatment group (high accessibility areas) at a particular time period (T1 or T2).\u003c/p\u003e \u003cp\u003e \u003cb\u003eWe used a parametric model to obtain standard errors and significance levels for the DID estimate\u003c/b\u003e:\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eDID Analysis: Primary Effects of Facility Accessibility\u003c/h2\u003e \u003cp\u003eFirst, we employed a general DID model to assess how different levels of facility accessibility (parks, public sports facilities, and commercial facilities) influenced PA duration. Given the distinct distribution patterns of facilities (park numbers: range 0\u0026ndash;35, mean 5.23, SD 4.775; public facilities: range 0\u0026ndash;4, median 1) and the few variations in park accessibility over the study period, we incorporated park numbers as fixed effects while classifying areas into high/low accessibility levels for public and commercial facilities at baseline. This specification demonstrated superior model fit.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eYit\u0026thinsp;=\u0026thinsp;β0\u0026thinsp;+\u0026thinsp;β1(Park\u003csub\u003ei\u003c/sub\u003e \u0026times; Time\u003csub\u003et\u003c/sub\u003e) + β2(Pub\u003csub\u003ei\u003c/sub\u003e\u0026times;Time\u003csub\u003et\u003c/sub\u003e) + β3(Com\u003csub\u003ei\u003c/sub\u003e\u0026times;Time\u003csub\u003et\u003c/sub\u003e) + αpark_j\u0026thinsp;+\u0026thinsp;γXit\u0026thinsp;+\u0026thinsp;εit\u003c/h2\u003e \u003cp\u003eWhere:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e\u0026bull; \u003cem\u003ePark\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e, \u003cem\u003epub\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e and \u003cem\u003eCom\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e indicate accessibility levels (high\u0026thinsp;=\u0026thinsp;1, low\u0026thinsp;=\u0026thinsp;0) for parks, public sports facilities, and commercial facilities respectively\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e\u0026bull; \u003cem\u003eβ0\u003c/em\u003e represents the baseline average\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e\u0026bull; \u003cem\u003eβ1\u003c/em\u003e -\u003cem\u003eβ3\u003c/em\u003e are the DID estimators that measure the differential changes in PA over time between areas with varying facility accessibility levels\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eDID Analysis: Commercial Facility Change Effects\u003c/h2\u003e \u003cp\u003eWe then focused on how changes in commercial facility accessibility affected PA duration:\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eYit\u0026thinsp;=\u0026thinsp;β0\u0026thinsp;+\u0026thinsp;β1 (Change\u003csub\u003ei\u003c/sub\u003e \u0026times; Time\u003csub\u003et\u003c/sub\u003e) + αpark_j\u0026thinsp;+\u0026thinsp;γXit\u0026thinsp;+\u0026thinsp;εit\u003c/h2\u003e \u003cp\u003eWhere \u003cem\u003eChange\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e indicates the change of the commercial facilities (stable/increased\u0026thinsp;=\u0026thinsp;1, decreased\u0026thinsp;=\u0026thinsp;0).\u003c/p\u003e \u003cp\u003eWe conducted this analysis for both the full sample and subsamples stratified by original commercial facility accessibility to examine potential heterogeneous effects.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eDDD Analysis: Public Facility Buffer Effects\u003c/h2\u003e \u003cp\u003eWe employed a DDD model to examine how public sports facilities moderated the relationship between commercial facility changes and PA duration, particularly focusing on areas with either low baseline accessibility or decreasing trends in commercial facilities:\u003c/p\u003e \u003cp\u003e \u003cem\u003eY\u003c/em\u003e \u003csub\u003e \u003cem\u003eit\u003c/em\u003e \u003c/sub\u003e\u0026thinsp;\u003cem\u003e=\u0026thinsp;β0\u0026thinsp;+\u0026thinsp;β1 (Pub\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e\u0026times; Com_disadv\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e \u003cem\u003e\u0026times; Time\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e) + β2 (Pub\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e \u003cem\u003e\u0026times; Time\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e) + β3 (Com_disadv\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e \u003cem\u003e\u0026times; Time\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e) + β4 (Pub\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e \u003cem\u003e\u0026times; Com\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e) + αpark_j\u0026thinsp;+\u0026thinsp;γXit\u0026thinsp;+\u0026thinsp;εit\u003c/em\u003e\u003c/p\u003e \u003cp\u003eWhere:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e\u0026bull; \u003cem\u003ePub\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e indicates public sports facility accessibility (high\u0026thinsp;=\u0026thinsp;1, low\u0026thinsp;=\u0026thinsp;0)\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e\u0026bull; \u003cem\u003eCom_disadv\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e represents disadvantaged areas in terms of commercial facilities (high baseline accessibility and stable/increased trend\u0026thinsp;=\u0026thinsp;0, others\u0026thinsp;=\u0026thinsp;1)\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eDID Analysis: Public Facility Buffer Effects in Different Commercial Facility Areas\u003c/h2\u003e \u003cp\u003eWe further investigated the buffering role of public facilities in areas with different commercial facility contexts by stratifying the sample into two groups: areas with commercial facility disadvantage (Com_disadv\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e=1) and areas with commercial facility advantage (Com_disadv\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e=0). We then conducted a stratified analysis using the following model:\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eYit\u0026thinsp;=\u0026thinsp;β0\u0026thinsp;+\u0026thinsp;β1 (Pub\u003csub\u003ei\u003c/sub\u003e\u0026times; Time\u003csub\u003et\u003c/sub\u003e) + αpark_j\u0026thinsp;+\u0026thinsp;γXit\u0026thinsp;+\u0026thinsp;εit\u003c/h2\u003e \u003cp\u003eWhere \u003cem\u003ePub\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e indicates public sports facility accessibility (high\u0026thinsp;=\u0026thinsp;1, low\u0026thinsp;=\u0026thinsp;0).\u003c/p\u003e \u003cp\u003eThis stratified analysis allowed us to assess whether the impact of public facilities on PA declines differed based on the pre-existing commercial facility environment and its changing trajectory during the pandemic.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eModel Validation\u003c/h2\u003e \u003cp\u003eThe Robustness analysis was performed adjusting control variables to test the robustness of the results. All analyses were conducted using STATA 17.0 (StataCorp, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), Statistical significance was set at p\u0026thinsp;\u0026lt;\u0026thinsp;0.05. We assessed multicollinearity among independent variables before regression analysis. Variance inflation factors (VIF) for all predictors remained below 5 (See Supplementary Table\u0026nbsp;1, Additional File 1), and Pearson correlation coefficients showed no values exceeding 0.7 (See Supplementary Table\u0026nbsp;2, Additional File 1), confirming acceptable independence among variables.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eDescriptive Statistics\u003c/h2\u003e \u003cp\u003eThe study sample comprised 701 participants with a balanced sex distribution (50% each). The majority of participants (67%) were aged 26\u0026ndash;45 years, with 69% holding a bachelor's degree or higher. Sixty percent reported annual household incomes between 120,000-360,000 CNY. The study areas were characterized by mean road network density of 0.025 (SD\u0026thinsp;=\u0026thinsp;0.005), land-use mix index of 0.701 (SD\u0026thinsp;=\u0026thinsp;0.063), and building density index of 0.205 (SD\u0026thinsp;=\u0026thinsp;0.067). Detailed demographic characteristics and environmental indicators are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eSpatial analysis revealed distinct distribution patterns of different facility types across neighborhoods. Commercial sports facilities, predominantly clustered in central urban areas with high socioeconomic status, experienced a substantial reduction from 2019 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea) to 2020 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb), with mean density decreasing from 70.194 (SD\u0026thinsp;=\u0026thinsp;111.576) to 35.578 (SD\u0026thinsp;=\u0026thinsp;21.857) facilities per catchment area. This change exhibited marked polarization: despite decreasing from 112to 45.8 facilities, high commercial facility accessibility areas maintained levels above the threshold of 43 facilities, while low commercial facility accessibility areas decreased from 31.033 to 25.983, consistently remaining below the threshold. In contrast, public sports facilities (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed) demonstrated more balanced and stable spatial distribution, maintaining a mean density of 1.02 (SD\u0026thinsp;=\u0026thinsp;1.171), supplemented by parks (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef) with a density of 5.233 (SD\u0026thinsp;=\u0026thinsp;4.775). This spatial configuration forms the foundation for examining how different facility types influenced residents' PA recovery patterns.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFollowing the reopening of sports facilities, participants' PA duration showed varying recovery trends. Overall (See Supplementary Table\u0026nbsp;3, Additional File 1), mean weekly LPA increased from 73.172 to 84.711 minutes, MVPA from 57.803 to 68.266 minutes, and total PA duration from 130.975 to 152.977 minutes per week. Spatial analysis revealed differential recovery patterns based on facility accessibility. People who live in areas with high park accessibility (\u0026ge;\u0026thinsp;4 parks) demonstrated not only recovery but exceeded T1 LPA levels (from 80.797 to 116.799 minutes/week), while low park accessibility areas (\u0026lt;\u0026thinsp;4 parks) showed decreased levels (from 68.308 to 64.243 minutes/week). Similarly, areas with high public facility accessibility exhibited increased MVPA duration (from 60.656 to 88.929 minutes/week), whereas low-accessibility areas experienced a decline (from 54.265 to 42.652 minutes/week). Areas with stable commercial facilities showed enhanced MVPA levels (from 62.535 to 93.655 minutes/week), while areas with decreased facilities remained below T1 levels (from 54.693 to 51.581 minutes/week). following commercial facility changes (T2), areas with high public facility accessibility and advantage commercial facilities maintained their elevated MVPA levels (from 86.896 to 88.63 minutes/week). Conversely, areas with both low public facility accessibility and disadvantaged accessibility to commercial facilities (n\u0026thinsp;=\u0026thinsp;81) experienced a further decrease in MVPA, from 54.046 to 34.282 minutes/week.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSample Characteristics and Environmental Features\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCharacteristic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003en (%) or Mean(SD)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDemographic characteristics\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (years)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e18\u0026ndash;25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e132 (19.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e26\u0026ndash;35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e261 (37.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e36\u0026ndash;45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e213 (30.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e46\u0026ndash;55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e82 (12.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026ge;\u0026thinsp;56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13 (2.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e349 (50.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e352 (50.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational attainment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBelow bachelor's degree\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e219 (31.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBachelor's degree or above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e482 (69.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAnnual household income (CNY)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;120,000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e123 (18.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e120,000-240,000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e221 (32.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e240,000-360,000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e199 (28.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e360,000-480,000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e44 (6.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e480,000-600,000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e52 (7.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;600,000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e62 (9.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eEnvironmental characteristics\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRoad network density\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.025 (0.005)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLand-use mix index\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.701 (0.063)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBuilding density index\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.205 (0.067)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePark density\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5.233 (4.775)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePublic sports facility density\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.02 (1.171)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCommercial sports facility density\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal T1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e70.194 (111.576)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e35.578 (21.857)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHigh accessibility T1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e112.012 (149.307)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e45.823 (24.58)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLow accessibility T1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e31.033 (9.352)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e25.983 (13.015)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003eEffects of Facilities Accessibility on PA duration\u003c/h2\u003e \u003cp\u003eOur difference-in-differences (DID) analyses revealed differential effects of facility accessibility on physical activity patterns. As shown in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, high park accessibility increased LPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;39.448, p\u0026thinsp;=\u0026thinsp;0.032) and total PA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;53.747, p\u0026thinsp;=\u0026thinsp;0.020). Areas with high accessibility to public sports facilities showed enhanced MVPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;39.069, p\u0026thinsp;=\u0026thinsp;0.015) and increased total physical activity duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;48.791, p\u0026thinsp;=\u0026thinsp;0.032). However, commercial facility accessibility did not significantly influence PA duration.\u003c/p\u003e \u003cp\u003eAmong built environment characteristics, building density showed a positive association with total physical activity duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;301.272, p\u0026thinsp;=\u0026thinsp;0.015), while land-use mix index was negatively associated with MVPA duration (\u003cem\u003eβ\u003c/em\u003e=-128.333, p\u0026thinsp;=\u0026thinsp;0.061). Regarding individual characteristics factors, age showed a positive association with LPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;26.867, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) but a negative association with MVPA duration (\u003cem\u003eβ\u003c/em\u003e=-11.734, p\u0026thinsp;=\u0026thinsp;0.008). Additionally, age was positively associated with total PA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;15.134, p\u0026thinsp;=\u0026thinsp;0.016). Compared to males, females demonstrated lower MVPA duration (\u003cem\u003eβ\u003c/em\u003e=-31.985, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Higher educational attainment was positively associated with MVPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;24.674, p\u0026thinsp;=\u0026thinsp;0.011), while higher annual household income was associated with increased LPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;7.338, p\u0026thinsp;=\u0026thinsp;0.028).\u003c/p\u003e \u003cp\u003eThe non-significant coefficients for both public sports facilities (LPA: \u003cem\u003eβ\u003c/em\u003e=-14.393, p\u0026thinsp;=\u0026thinsp;0.297; MVPA: \u003cem\u003eβ\u003c/em\u003e=-5.133, p\u0026thinsp;=\u0026thinsp;0.677; Total PA: \u003cem\u003eβ\u003c/em\u003e=-19.526, p\u0026thinsp;=\u0026thinsp;0.263) and commercial sports facilities (LPA: \u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;6.909, p\u0026thinsp;=\u0026thinsp;0.621; MVPA: \u003cem\u003eβ\u003c/em\u003e=-8.279, p\u0026thinsp;=\u0026thinsp;0.507; Total PA: \u003cem\u003eβ\u003c/em\u003e=-1.369, p\u0026thinsp;=\u0026thinsp;0.938) support the parallel trends assumption, suggesting that pre-intervention physical activity patterns were similar across areas with different facility accessibility.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEffects of facility accessibility on PA duration: DID analysis\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLPA duration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMVPA duration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTotal PA duration\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eβ(P-value)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eβ (P-value)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eβ (P-value)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDID (\u003cem\u003ePark\u003c/em\u003e \u003cem\u003e\u0026times; Time\u003c/em\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e39.448(0.032*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14.299(0.382)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e53.747(0.020*)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDID (\u003cem\u003ePub \u0026times; Time\u003c/em\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9.721(0.590)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e39.069(0.015*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48.791(0.032*)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDID (\u003cem\u003ecom \u0026times; Time\u003c/em\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-8.843(0.621)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.759(0.814)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-5.084(0.822)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePublic facilities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-14.393(0.297)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-5.133(0.677)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-19.526(0.263)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCommercial facilities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6.909(0.621)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-8.279(0.507)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-1.369(0.938)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-4.915(0.778)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-18.553(0.235)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-23.468(0.288)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRoad network density\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-2242.578(0.131)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-365.269(0.783)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-2607.846(0.164)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBuilding density index\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e166.041(0.089)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e135.230(0.121)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e301.272(0.015**)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLand-use mix index\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8.143(0.915)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-128.333(0.061*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-120.190(0.214)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational attainment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-18.764(0.083*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24.674(0.011**)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.909(0.665)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-11.357(0.211)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-31.985(0.000***)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-43.342(0.000***)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e26.867(0.000***)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-11.734(0.008***)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e15.134(0.016**)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAnnual household income\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7.338(0.028*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.556(0.852)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.891(0.061*)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e46.285(0.482)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e137.136(0.020*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e183.421(0.028*)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u0026sup2;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.083\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0792\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0807\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1402\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1402\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1402\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001. N\u0026thinsp;=\u0026thinsp;1,402 observations (701 individuals \u0026times; 2 time points)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003eEffects of Commercial Sports Facilities Accessibility Changes on PA duration:\u003c/h2\u003e \u003cp\u003eAs shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, in the full sample, the stability of commercial facilities positively influenced both MVPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;41.033, p\u0026thinsp;=\u0026thinsp;0.001) and total PA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;38.181, p\u0026thinsp;=\u0026thinsp;0.032). These effects exhibited distinct patterns across neighborhoods with different baseline accessibility levels.\u003c/p\u003e \u003cp\u003eThe impact was particularly pronounced in low commercial facility accessibility neighborhoods, which is attributed to the stability of commercial facilities leading to a substantial increase in MVPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;70.019, p\u0026thinsp;=\u0026thinsp;0.000) and total PA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;77.406, p\u0026thinsp;=\u0026thinsp;0.004). In contrast, high commercial facility accessibility neighborhoods showed minimal response to commercial facility changes (MVPA: \u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;19.708, p\u0026thinsp;=\u0026thinsp;0.252; total PA: \u003cem\u003eβ\u003c/em\u003e=-0.179, p\u0026thinsp;=\u0026thinsp;0.994), suggesting a diminishing marginal effect of additional commercial facilities in areas with adequate existing resources.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEffects of Commercial Facility Changes on PA duration: Stratified DID Analysis\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003ePanel A: Full Sample Analysis\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLPA duration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMVPA duration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTotal PA duration\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eβ(P-value)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eβ (P-value)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eβ (P-value)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDID(\u003cem\u003eChange\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e\u0026times; Time\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-2.853(0.839)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e41.033(0.001***)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e38.181(0.032*)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePublic facilities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-9.359(0.372)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11.899(0.203)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.541(0.848)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCommercial facilities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.112(0.846)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.991(0.919)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.121(0.935)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12.670(0.228)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5.809(0.535)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.861(0.606)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eControl Variables\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u0026sup2;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0794\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0816\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0764\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1,402\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1,402\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1,402\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePanel B: Stratified Analysis by Baseline Accessibility\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLow commercial facility accessibility\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDID (\u003cem\u003eChange\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e \u003cem\u003e\u0026times; Time\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7.387(0.723)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e70.019(0.000***)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e77.406(0.004***)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePublic facilities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20.039(0.308)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.363(0.847)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e23.402(0.349)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12.221(0.464)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-24.188(0.103)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-11.966(0.573)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eControl Variables\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u0026sup2;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0779\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.1247\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.1028\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e722\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e722\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e722\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eHigh commercial facility accessibility\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDID (\u003cem\u003eChange\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e \u003cem\u003e\u0026times; Time\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-19.886(0.314)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e19.708(0.252)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.179(0.994)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePublic facilities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-39.415(0.006**)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22.542(0.070*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-16.873(0.338)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12.874(0.328)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.950(0.665)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17.825(0.272)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eControl Variables\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u0026sup2;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.1401\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.1293\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.1259\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e680\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e680\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e680\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003eNote: Control Variables include built environment and individual characteristics.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001. Control Variables include built environment features (population density, land-use mix, street connectivity), individual characteristics (age, gender, education, income), and park-specific fixed effects\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThis polarizing effect was further evidenced by neighborhood-specific built environment characteristics. In high-accessibility neighborhoods, the influence of commercial facility changes was overshadowed by existing built environment features, particularly building density (total PA: \u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;719.939, p\u0026thinsp;=\u0026thinsp;0.000). These findings confirm our hypothesis that the effectiveness of commercial facility changes is contingent upon the baseline accessibility level, with areas of limited commercial resources benefiting most from such interventions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003eJoint Effects of Commercial and Public Sports Facilities on PA duration:\u003c/h2\u003e \u003cp\u003eThe DDD analysis revealed patterns in how public sports facilities moderate physical activity (PA) inequities. As shown in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, the triple-interaction term showed a positive association with MVPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;79.269, p\u0026thinsp;=\u0026thinsp;0.050), indicating that public facilities effectively buffer against MVPA inequality in commercially disadvantaged areas. This positive coefficient suggests that public facilities have stronger positive effects on MVPA in neighborhoods with commercial facility disadvantages (either low baseline accessibility or decreasing trends) compared to commercially advantaged areas. Notably, no significant interaction between facility types was observed prior to facility changes (\u003cem\u003eβ\u003c/em\u003e=-31.337, p\u0026thinsp;=\u0026thinsp;0.168), suggesting that the equity-promoting effect of public facilities emerged specifically in response to commercial facility changes. The effectiveness of this buffering mechanism was further contextualized by both built environment and individual characteristics. Among built environment factors, building density showed positive associations with total PA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;305.468, p\u0026thinsp;=\u0026thinsp;0.014). Individual characteristics factors also played significant roles, with educational attainment (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;23.599, p\u0026thinsp;=\u0026thinsp;0.015) showing a positive association, while both female gender (\u003cem\u003eβ\u003c/em\u003e=-31.200, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and age (\u003cem\u003eβ\u003c/em\u003e=-11.878, p\u0026thinsp;=\u0026thinsp;0.008) were negatively associated with MVPA duration. Annual household income showed a positive association with LPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;6.991, p\u0026thinsp;=\u0026thinsp;0.037) but not with MVPA specifically (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.897, p\u0026thinsp;=\u0026thinsp;0.763).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eJoint Effects of Public and Commercial Facilities: Triple-Difference Analysis\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLPA duration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMVPA duration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTotal PA duration\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eβ(P-value)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eβ (P-value)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eβ (P-value)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDDD (\u003cem\u003ePub\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e\u0026times; Com_disadv\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e \u003cem\u003e\u0026times; Time\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-19.575(0.667)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e79.269(0.050*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e59.694(0.299)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePub\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e\u0026times; Time\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e31.449(0.464)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-28.546(0.456)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.902(0.957)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCom_disadv\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e \u003cem\u003e\u0026times; Time\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e34.905(0.269)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-47.683(0.009**)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-12.777(0.749)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePub\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e\u0026times; Com_disadv\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10.510(0.680)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-31.337(0.168)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-20.827(0.518)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePublic facilities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-24.545(0.326)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20.264(0.363)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-4.281(0.892)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCommercial facilities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7.105(0.538)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-10.738(0.296)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.632(0.803)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-26.690(0.386)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.280(0.269)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.590(0.926)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRoad network density\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-1957.725(0.192)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-714.726(0.592)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-2672.448(0.158)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBuilding density index\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e160.960(0.101)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e144.508(0.099)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e305.468(0.014**)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLand-use mix index\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-10.442(0.896)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-112.880(0.111)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-123.322(0.220)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational attainment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-17.748(0.102)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.599(0.015**)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.852(0.669)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-12.028(0.187)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-31.200(0.000***)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-43.228(0.000***)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e27.038(0.000***)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-11.878(0.008***)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e15.160(0.016**)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAnnual Household Income\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6.991(0.037*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.897(0.763)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.884(0.062*)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e53.046(0.424)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e135.660(0.022*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e188.707(0.025*)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u0026sup2;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0813\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0818\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.078\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1402\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1402\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1402\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFurther analysis stratified by commercial facility advantage (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e) revealed that while public facilities positively promoted MVPA in both commercially advantaged and disadvantaged areas, the effect was significant only in commercially disadvantaged neighborhoods (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;42.506, p\u0026thinsp;=\u0026thinsp;0.001). In contrast, commercially advantaged areas showed a positive but non-significant effect (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;6.464, p\u0026thinsp;=\u0026thinsp;0.868). This finding suggests that public facilities play a crucial role in mitigating MVPA inequalities, particularly in areas experiencing commercial facility disadvantages.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEffects of Public Sports Facilities on PA Duration by Commercial Facility Advantage\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eDisadvantage Commercial Facilities\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003eAdvantage Commercial Facilities\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;601)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;100)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLPA duration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMVPA duration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTotal PA duration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLPA duration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMVPA duration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eTotal PA duration\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDID(\u003cem\u003eChange \u0026times; Time\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-3.261(0.827)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e42.506(0.001***)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e39.245(0.034*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-13.372(0.677)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.464(0.868)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-6.909(0.886)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16.267(0.202)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-14.026(0.200)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.242(0.887)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.041(0.941)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e15.067(0.648)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e17.108(0.677)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eControl Variables\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u0026sup2;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0752\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0798\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0827\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.3085\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.1851\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.2103\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e \u003cp\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001. Control Variables include built environment features (population density, land-use mix, street connectivity), individual characteristics (age, gender, education, income), and park-specific fixed effects\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThese findings provide empirical evidence supporting the strategic placement of public sports facilities as an intervention tool to promote PA equity, particularly in areas experiencing commercial facility disadvantages.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003eRobustness tests\u003c/h2\u003e \u003cp\u003eThe robustness checks of our findings through progressive addition of control variables demonstrated consistent results (Wang et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). As shown in Supplementary Table\u0026nbsp;4, Additional File 1, our results demonstrate strong stability across different model specifications. Specifically, park accessibility shows consistent positive effects on LPA (39.45\u0026ndash;51.86 minutes/week, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). While public sports facilities show no significant impact on LPA, they demonstrate stable positive effects on MVPA (31.17\u0026ndash;40.94 minutes/week, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01) and total physical activity (26.81\u0026ndash;48.79 minutes/week, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e \u003cp\u003eThe effects of commercial facility changes exhibit spatial heterogeneity: in low-accessibility areas, they promote both MVPA (42.67\u0026ndash;70.02 minutes/week, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01) and total physical activity (68.35-88.00 minutes/week, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), while showing diminishing effects in high-accessibility areas (from 33.79 minutes/week, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 to 19.71 minutes/week, p\u0026thinsp;\u0026gt;\u0026thinsp;0.05). The interaction between commercial disadvantage, public facilities, and time (DDD) reveals a positive impact on MVPA (71.20-89.96 minutes/week, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), indicating that public sports facilities have stronger effects in commercially disadvantaged areas over time. This finding is further supported by our stratified analysis, which shows that public facilities have significant positive effects on MVPA (42.51\u0026ndash;49.14 minutes/week, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) in commercially disadvantaged areas, while showing no significant effects in commercially advantaged areas. These estimates remain robust after controlling for individual characteristics, built environment features, and park fixed effects.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec30\" class=\"Section2\"\u003e \u003ch2\u003eMain findings\u003c/h2\u003e \u003cp\u003eThis study examined how different levels of neighborhood facility accessibility influenced residents' PA patterns during COVID-19 facility closures and reopening. Public facilities (parks and public sports facilities) demonstrated resilience with stable accessibility levels, while commercial facilities exhibited market-driven fluctuations, reflecting different operational mechanisms during the crisis. Facility changes affected PA recovery differently.(Larson et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Our analysis revealed three key findings:\u003c/p\u003e \u003cp\u003eFirst, our study confirms that higher facility accessibility is associated with increased PA duration. Specifically, in areas with high facility accessibility, participants' PA levels not only recovered but also exceeded pre-pandemic levels following the reopening of facilities. This finding aligns with previous research documenting PA recovery trends(Ding et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Park et al., n.d.).For example, DID analysis revealed that areas with high park accessibility experienced greater increases in LPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;39.448, p\u0026thinsp;=\u0026thinsp;0.032) and total PA (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;53.747, p\u0026thinsp;=\u0026thinsp;0.020) compared to low-accessibility areas(Hoekman et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).Similarly, high public sports facility accessibility was associated with greater MVPA duration increases (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;39.069, p\u0026thinsp;=\u0026thinsp;0.015) and total physical activity (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;48.791, p\u0026thinsp;=\u0026thinsp;0.032). This functional differentiation supports theories on the multifaceted value of public facilities (Bergsgard et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) and confirms their fundamental role in promoting equitable PA recovery (Jones et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). However, baseline commercial facility accessibility showed no significant association with PA changes (LPA: \u003cem\u003eβ\u003c/em\u003e=-8.843, p\u0026thinsp;=\u0026thinsp;0.621; MVPA: \u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.759, p\u0026thinsp;=\u0026thinsp;0.814).This non-significant relationship between commercial facility accessibility and PA changes may be attributed to market saturation diminishing marginal returns, diverse exercise options in high-accessibility areas diluting impact, and facility quantity not reflecting quality or utilization patterns.(M\u0026uuml;ller et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) .\u003c/p\u003e \u003cp\u003eSecond, commercial facilities exhibited distinct market-oriented spatial patterns during the pandemic, with pronounced polarization effects: while high-accessibility areas maintained sufficient facility levels despite overall decreases, low commercial facility accessibility areas experienced further reductions in accessibility. (Shen et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This spatial polarization manifested in differential impacts on MVPA recovery: residents in low-accessibility areas showed sensitivity to facility stable (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;70.019, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), whereas those in high-accessibility areas demonstrated no significant association (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;19.708, p\u0026thinsp;=\u0026thinsp;0.252) (Powell et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2006\u003c/span\u003e).This market-driven uneven distribution of commercial facilities not only exacerbated existing inequalities in PA recovery but also potentially limited facility investment in resource-deprived areas through the \"Matthew Effect\" (Rivera-Navarro et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; van Lenthe et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), consistent with previous findings on health resource spatial aggregation (Xu et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThird, our study pioneered the quantification of public facilities' buffering effect through triple-difference analysis, revealing that public facilities can significantly buffer the negative impact of commercial facility disadvantages on MVPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;79.269, p\u0026thinsp;=\u0026thinsp;0.050). This positive coefficient demonstrates that within the context of neighborhood deprivation, public facilities effectively mitigate the adverse effects of commercial facility disadvantages on physical activity patterns. Stratified analysis confirmed that public facilities positively promoted MVPA across all areas, but the effect was significant only in commercially disadvantaged regions (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;42.506, p\u0026thinsp;=\u0026thinsp;0.001). This pattern reveals a conditional buffering effect: public facilities provide substantial protection against physical activity inequality in deprived neighborhoods while offering diminishing marginal returns in commercially advantaged environments. In areas with high public facility accessibility, residents maintained and even increased their MVPA levels (from 55.666 to 88.986 minutes/week) despite commercial facility disadvantages (Chastin et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) .(White et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). This buffering effect, particularly evident during market fluctuation, suggests that strategic public facility placement could serve as an effective intervention tool for reducing health inequities. (Milton et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Rogers et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNotably, areas with high public facility accessibility showed lower total PA duration despite supporting MVPA, consistent with previous findings(Yang et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). This counterintuitive result may be explained by residents' engagement in time-limited, structured activities rather than sustained LPA, particularly following the post-pandemic shift toward indoor exercise (Sun et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec31\" class=\"Section2\"\u003e \u003ch2\u003ePolicy Implications and Practical Significance\u003c/h2\u003e \u003cp\u003eBased on our research findings, we propose three specific policy recommendations for improving urban PA facility provision:\u003c/p\u003e \u003cp\u003eFirst, policymakers should continue integrating public sports facilities into basic public services, especially in areas with commercial facility disadvantages. Our study emphasizes the fundamental role of public PA facilities in promoting PA participation and their buffering effect against neighborhood health deprivation (Powell et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). This finding aligns with public health approaches emphasizing structural interventions to address health inequities rather than focusing solely on individual behavior change (H\u0026oslash;yer-Kruse et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Urban planning departments should: Incorporate PA facilities into basic livelihood security, with particular attention to facility allocation in deprived neighborhoods (Reijneveld et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Rivera-Navarro et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2022\u003c/span\u003e);For areas with limited fiscal resources, encourage the shared use of existing facilities such as school sports venues. This approach can provide affordable public facilities while optimizing resource allocation (Hoekman et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Luo, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Importantly, even in areas with high commercial facility density, public sports facilities still showed a positive effect, though not statistically significant. Deploying public sports facilities remains valuable in these areas, as low-income residents may be excluded from physical activity opportunities due to economic barriers to accessing commercial facilities(Higgerson et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). This strategic deployment ensures physical activity opportunities for all socioeconomic groups and prevents intra-neighborhood health inequalities(SFM et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSecond, urban planning departments should optimize the functional layout of PA facilities based on their differentiated effects. Our research reveals that different facility types influence specific dimensions of residents' PA behavior: parks and green spaces primarily promote LPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;39.448, p\u0026thinsp;=\u0026thinsp;0.032) (Iamtrakul et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), while professional sports facilities are more conducive to increasing MVPA duration(\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;39.069, p\u0026thinsp;=\u0026thinsp;0.015)(Jones et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). To meet residents' diverse PA needs(Eime et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), we recommend incorporating professional exercise zones within parks and creating integrated facility networks that serve both casual and structured exercise purposes. This functional differentiation approach can maximize land use efficiency while addressing the full spectrum of PA needs.\u003c/p\u003e \u003cp\u003eThird, local governments should develop collaborative management models that address the spatial polarization of PA resources. Our finding that commercial facilities show market-driven distribution patterns with diminishing returns in high-accessibility areas (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;19.708, p\u0026thinsp;=\u0026thinsp;0.252) versus significant effects in low-accessibility areas (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;70.019, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) suggests the need for targeted interventions: Implement conditional incentive policies (e.g., venue subsidies, tax benefits) specifically for commercial facility development in underserved areas, thereby improving facility accessibility and reducing spatial inequality(Humphreys \u0026amp; Zhou, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Foster public-private partnerships through third-party organizations that can: Facilitate shared use agreements between sports clubs and public venues; Develop needs-based compensation mechanisms for facility providers in disadvantaged areas; Coordinate facility scheduling and maintenance to maximize utilization in areas with limited resources. This collaborative approach can enhance facility utilization while ensuring operational sustainability and spatial equity (Kang \u0026amp; Lee, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Kenyon et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec32\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eSeveral limitations should be noted. First, our reliance on self-reported PA data may introduce recall bias, particularly given the retrospective nature of pandemic-related data collection. Second, despite using DID analysis to control for time-fixed effects, self-selection bias might exist where residents with stronger PA preferences choose neighborhoods with better facility access. Third, while our study in Shenzhen provides valuable insights, findings may not fully generalize to other urban contexts with different socioeconomic and built environment characteristics. Finally, our focus on facility accessibility does not capture other aspects such as facility quality and programming that may influence PA participation.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThrough a natural experiment of facility closures during COVID-19, this study employed DID and DDD analyses to examine how different types of sports facilities influence PA duration. We found distinct patterns between public and commercial facilities. Commercial facilities showed market-driven spatial clustering with differential impacts on MVPA in high-accessibility (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;19.708, p\u0026thinsp;=\u0026thinsp;0.252) versus low-accessibility areas (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;70.019, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), suggesting diminishing marginal returns that potentially exacerbate health inequities. Public facilities demonstrated more balanced distribution and complementary effects: parks facilitated LPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;39.448, p\u0026thinsp;=\u0026thinsp;0.032) and public sports facilities enhanced MVPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;39.069, p\u0026thinsp;=\u0026thinsp;0.015). Through DDD analysis, we found that public sports facilities effectively buffered the negative impact of commercial facility disadvantages on MVPA duration (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;79.269, p\u0026thinsp;=\u0026thinsp;0.050), with this buffering effect being particularly significant in neighborhoods experiencing commercial facility deprivation (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;42.506, p\u0026thinsp;=\u0026thinsp;0.001). These findings provide evidence-based implications for addressing health inequities through environmental interventions and inform policies aimed at building resilient and equitable PA environments, particularly highlighting the value of strategic public facility investment in commercially disadvantaged neighborhoods.Our study identifies built environments as key social determinants of health, with public facilities' buffering effects supporting structural public health approaches to improve population-level physical activity and health outcomes.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by the Institutional Research Ethics Board at Harbin Institute of Technology. All participants were fully informed about the design and the purpose of the study, and they provided informed consent before participating. The research was conducted in accordance with the principles of the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDue to privacy and confidentiality agreements, the raw dataset from this study is not publicly available. However, aggregated data can be made available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Natural Science Foundation of China [grant number 42371238], the Guangdong Province Philosophy and Social Sciences Planning Project [grant number GD23XGL087], and the Shenzhen Science and Technology Innovation Commission [grant number 0231129125223001]. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePZ conceptualized the study idea, provided methodological guidance, data resources, and critically reviewed and revised the manuscript. JY performed data analysis and wrote the original draft of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe would like to express our incere thanks to Yirou Chen, Xinsu Lv, and ZhenHu for their dedication and assistance during the fieldwork and data collection.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Natural Science Foundation of China under Grant [number 42371238], the Guangdong Province Philosophy and Social Sciences Planning Project [number GD23XGL087], and the Shenzhen Science and Technology Innovation Commission under Grant [number 0231129125223001].\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAinsworth BE. (2024). \u003cem\u003e2024 Adult Compendium \u0026ndash; Compendium of Physical Activities\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pacompendium.com/adult-compendium/\u003c/span\u003e\u003cspan address=\"https://pacompendium.com/adult-compendium/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBergsgard NA, Borodulin K, Fahlen J, Hoyer-Kruse J, Iversen EB. National structures for building and managing sport facilities: A comparative analysis of the Nordic countries. SPORT Soc. 2019;22(4):525\u0026ndash;39. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/17430437.2017.1389023\u003c/span\u003e\u003cspan address=\"10.1080/17430437.2017.1389023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChastin SFM, Van Cauwenberg J, Maenhout L, Cardon G, Lambert E, Van Dyck D. Inequality in physical activity, global trends by income inequality and gender in adults. Int J Behav Nutr Phys ACTIVITY. 2020;17(1):142. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12966-020-01039-x\u003c/span\u003e\u003cspan address=\"10.1186/s12966-020-01039-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCulture R. Television, Tourism andSports Bureau of Shenzhen Municipality. (2019). \u003cem\u003eList of Parks in Shenzhen\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://cgj.sz.gov.cn/gkmlpt/content/5/5402/post_5402966.html#476\u003c/span\u003e\u003cspan address=\"https://cgj.sz.gov.cn/gkmlpt/content/5/5402/post_5402966.html#476\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDing D, del Pozo Cruz B, Green MA, Bauman AE. Is the COVID-19 lockdown nudging people to be more active: A big data analysis. Br J Sports Med. 2020;54(20):1183\u0026ndash;1183. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bjsports-2020-102575\u003c/span\u003e\u003cspan address=\"10.1136/bjsports-2020-102575\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDur\u0026atilde;o S, Burns J, Schmidt B-M, Tumusiime D, Hohlfeld A, Pfadenhauer L, Ongolo-Zogo C, Rehfuess E, Kredo T. Infrastructure, policy and regulatory interventions to increase physical activity to prevent cardiovascular diseases and diabetes: A systematic review. BMC Public Health. 2023;23:112. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12889-022-14841-y\u003c/span\u003e\u003cspan address=\"10.1186/s12889-022-14841-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEime RM, Harvey J, Charity MJ, Casey M, Westerbeek H, Payne WR. The relationship of sport participation to provision of sports facilities and socioeconomic status: A geographical analysis. Aust N Z J Public Health. 2017;41(3):248\u0026ndash;55. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1753-6405.12647\u003c/span\u003e\u003cspan address=\"10.1111/1753-6405.12647\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEriksson U, Arvidsson D, Sundquist K. Availability of exercise facilities and physical activity in 2,037 adults: Cross-sectional results from the Swedish neighborhood and physical activity (SNAP) study. BMC Public Health. 2012;12:607. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/1471-2458-12-607\u003c/span\u003e\u003cspan address=\"10.1186/1471-2458-12-607\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEsri. (2020). ArcGIS Desktop: Release 10.8. Redlands, CA: Environmental Systems Research Institute[^121^]. \u003cem\u003eEsri\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.esri.com/about/newsroom/announcements/esri-releases-guide-teaching-arcgis-desktop-10-8/\u003c/span\u003e\u003cspan address=\"https://www.esri.com/about/newsroom/announcements/esri-releases-guide-teaching-arcgis-desktop-10-8/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGidlow C, Cerin E, Sugiyama T, Adams MA, Mitas J, Akram M, Reis RS, Davey R, Troelsen J, Schofield G, Sallis JF. Objectively measured access to recreational destinations and leisure-time physical activity: Associations and demographic moderators in a six-country study. Health Place. 2019;59:102196. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.healthplace.2019.102196\u003c/span\u003e\u003cspan address=\"10.1016/j.healthplace.2019.102196\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGomes TN, Suikkanen S, Thuany M, Vaananen I, Donnelly A. Physical activity in outdoor spaces, sleep quality and general health: A pilot feasibility study. Lancet (London England). 2023;402(Suppl 1):S45. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0140-6736(23)02094-9\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(23)02094-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHiggerson J, Halliday E, Ortiz-Nunez A, Brown R, Barr B. Impact of free access to leisure facilities and community outreach on inequalities in physical activity: A quasi-experimental study. J Epidemiol Commun Health. 2018;72(3):252\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/jech-2017-209882\u003c/span\u003e\u003cspan address=\"10.1136/jech-2017-209882\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoekman R, Grubben M, Kraaykamp G. (2023). Sport club dropout under COVID-19 in the Netherlands: Do characteristics of the neighbourhood matter? \u003cem\u003eFrontiers in Sports and Active Living\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fspor.2023.1168608\u003c/span\u003e\u003cspan address=\"10.3389/fspor.2023.1168608\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH\u0026oslash;yer-Kruse J, Schmidt EB, Hansen AF, Pedersen MRL. The interplay between social environment and opportunities for physical activity within the built environment: A scoping review. BMC Public Health. 2024;24(1):2361. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12889-024-19733-x\u003c/span\u003e\u003cspan address=\"10.1186/s12889-024-19733-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHumphreys BR, Zhou L. Sports facilities, agglomeration, and public subsidies. Reg Sci Urban Econ. 2015;54:60\u0026ndash;73. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.regsciurbeco.2015.07.004\u003c/span\u003e\u003cspan address=\"10.1016/j.regsciurbeco.2015.07.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIamtrakul P, Chayphong S, Gao W. Assessing spatial disparities and urban facility accessibility in promoting health and well-being. Transp Res Interdisciplinary Perspect. 2024;25:101126. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.trip.2024.101126\u003c/span\u003e\u003cspan address=\"10.1016/j.trip.2024.101126\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang Y, Liu Y, Liu Z, Wang C, Shi Z, Zhao H, Sun D, Sun W, Wang X. Spatial Distribution Characteristics of Public Fitness Venues: An Urban Accessibility Perspective. SUSTAINABILITY. 2023;15(1):601. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su15010601\u003c/span\u003e\u003cspan address=\"10.3390/su15010601\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJing Y, Liu Y, Cai E, Liu Y, Zhang Y. Quantifying the spatiality of urban leisure venues in Wuhan, Central China\u0026mdash;GIS-based spatial pattern metrics. SUSTAINABLE CITIES Soc. 2018;40:638\u0026ndash;47. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.scs.2018.05.015\u003c/span\u003e\u003cspan address=\"10.1016/j.scs.2018.05.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJones A, Hillsdon M, Coombes E. Greenspace access, use, and physical activity: Understanding the effects of area deprivation. Prev Med. 2009;49(6):500\u0026ndash;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ypmed.2009.10.012\u003c/span\u003e\u003cspan address=\"10.1016/j.ypmed.2009.10.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJowshan M-R, Pourjavid A, Amirkhizi F, Hosseini M-H, Zolghadrpour M-A, Hamedi-Shahraki S, Motlagh AH, Asghari S. Adherence to combined healthy lifestyle and odds of metabolic syndrome in Iranian adults: The PERSIAN Dena cohort study. Sci Rep. 2025;15(1):5164. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-025-89028-7\u003c/span\u003e\u003cspan address=\"10.1038/s41598-025-89028-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKahneman D, Krueger AB, Schkade DA, Schwarz N, Stone AA. A Survey Method for Characterizing Daily Life Experience: The Day Reconstruction Method. Science. 2004;306(5702):1776\u0026ndash;80. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.1103572\u003c/span\u003e\u003cspan address=\"10.1126/science.1103572\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKang M, Lee Y. The Gap in Community Sports: Utilization of Sports Facilities in South Korea. Int J Environ Res Public Health. 2022;19(8). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijerph19084495\u003c/span\u003e\u003cspan address=\"10.3390/ijerph19084495\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Article 8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKegler MC, Gauthreaux N, Hermstad A, Arriola KJ, Mickens A, Ditzel K, Hernandez C, Haardoerfer R. Inequities in Physical Activity Environments and Leisure-Time Physical Activity in Rural Communities. Prev Chronic Dis. 2022;19:210417. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5888/pcd19.210417\u003c/span\u003e\u003cspan address=\"10.5888/pcd19.210417\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKenyon JA, Mason C, Rookwood J. Emerging third-sector sports organisations and navigating uncertainty in an era of austerity: A single ethnographic case study from Liverpool. Int J SPORT POLICY POLITICS. 2018;10(1):25\u0026ndash;42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/19406940.2018.1425732\u003c/span\u003e\u003cspan address=\"10.1080/19406940.2018.1425732\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKohl HW, Craig CL, Lambert EV, Inoue S, Alkandari JR, Leetongin G, Kahlmeier S. The pandemic of physical inactivity: Global action for public health. Lancet. 2012;380(9838):294\u0026ndash;305. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0140-6736(12)60898-8\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(12)60898-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLarson LR, Zhang Z, Oh JI, Beam W, Ogletree S, Bocarro J, Lee K, Casper JM, Stevenson K, Hipp J, Mullenbach LE, Carusona M, Wells ML. Urban Park Use During the COVID-19 Pandemic: Are Socially Vulnerable Communities Disproportionately Impacted? Front Sustainable Cities. 2021. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/frsc.2021.710243\u003c/span\u003e\u003cspan address=\"10.3389/frsc.2021.710243\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee S, Lee C, Xu M, Li W, Ory M. People living in disadvantaged areas faced greater challenges in staying active and using recreational facilities during the COVID-19 pandemic. Health Place. 2022;75:102805. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.healthplace.2022.102805\u003c/span\u003e\u003cspan address=\"10.1016/j.healthplace.2022.102805\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi B, Valerino-Perea S, Zhou W, Xie Y, Syrett K, Peters R, He Z, Zou Y, de Vocht F, Foster C. The impact of the world\u0026rsquo;s first regulatory, multi-setting intervention on sedentary behaviour among children and adolescents (ENERGISE): A natural experiment evaluation. Int J Behav Nutr Phys ACTIVITY. 2024;21(1):53. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12966-024-01591-w\u003c/span\u003e\u003cspan address=\"10.1186/s12966-024-01591-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLisbona Guill\u0026eacute;n M, S\u0026aacute;nchez Mart\u0026iacute;n J, S\u0026aacute;nchez Mart\u0026iacute;n R. Urban segregation and body differentiation: Private gyms in Merida (Mexico). Iztapalapa Revista de ciencias sociales y humanidades. 2024;45(96):235\u0026ndash;63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.28928/ri/962024/aot6/lisbona/etal/x\u003c/span\u003e\u003cspan address=\"10.28928/ri/962024/aot6/lisbona/etal/x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu M, Timmermans EJ, Zou D, Grobbee DE, Zhou S, Vaartjes I. Impact of green space exposure on blood pressure in Guangzhou, China: Mediation by air pollution, mental health, physical activity, and weight status. Environ Pollut. 2024;356:124251. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.envpol.2024.124251\u003c/span\u003e\u003cspan address=\"10.1016/j.envpol.2024.124251\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu Y, Wang X, Zhou S, Wu W. The association between spatial access to physical activity facilities within home and workplace neighborhoods and time spent on physical activities: Evidence from Guangzhou, China. Int J Health Geogr. 2020;19(1):22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12942-020-00216-2\u003c/span\u003e\u003cspan address=\"10.1186/s12942-020-00216-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLundh H, Arvidsson D, Greven C, Fridolfsson J, B\u0026ouml;rjesson M, Boman C, Lauruschkus K, Lundqvist S, Melin K, Bernhardsson S. Physical activity and sedentary behaviour amongst children with obesity\u0026mdash;Exploring cross-sectional associations between child and parent. J Activity Sedentary Sleep Behav. 2025;4(1):2. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s44167-025-00072-0\u003c/span\u003e\u003cspan address=\"10.1186/s44167-025-00072-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuo Q. (2018). Research on the Predicament and Optimization of the Openness of College Sports Facilities in China. In Y. M. Wang \u0026amp; D. C. Zhou, editors, \u003cem\u003e2018 4TH INTERNATIONAL CONFERENCE ON EDUCATION, MANAGEMENT AND INFORMATION TECHNOLOGY (ICEMIT 2018)\u003c/em\u003e (pp. 992\u0026ndash;996). Francis Acad Press. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000450415100224\u003c/span\u003e\u003cspan address=\"https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:000450415100224\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMacdonald L. Associations between spatial access to physical activity facilities and frequency of physical activity; how do home and workplace neighbourhoods in West Central Scotland compare? Int J Health Geogr. 2019;18:2. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12942-019-0166-z\u003c/span\u003e\u003cspan address=\"10.1186/s12942-019-0166-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMenardo E, De Dominicis S, Pasini M. Exploring Perceived and Objective Measures of the Neighborhood Environment and Associations with Physical Activity among Adults: A Review and a Meta-Analytic Structural Equation Model. Int J Environ Res Public Health. 2022;19(5):2575. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijerph19052575\u003c/span\u003e\u003cspan address=\"10.3390/ijerph19052575\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMilton K, Cavill N, Chalkley A, Foster C, Gomersall S, Hagstromer M, Kelly P, Kolbe-Alexander T, Mair J, McLaughlin M, Nobles J, Reece L, Shilton T, Smith BJ, Schipperijn J. Eight Investments That Work for Physical Activity. J Phys Act Health. 2021;18(6):625\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1123/jpah.2021-0112\u003c/span\u003e\u003cspan address=\"10.1123/jpah.2021-0112\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM\u0026uuml;ller C, Paulsen L, Bucksch J, Wallmann-Sperlich B. Built and natural environment correlates of physical activity of adults living in rural areas: A systematic review. Int J Behav Nutr Phys Activity. 2024;21(1):52. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12966-024-01598-3\u003c/span\u003e\u003cspan address=\"10.1186/s12966-024-01598-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO\u0026rsquo;Brien D. Commercial sport organisations and community capacity building: A case study of two surf parks. SPORT Manage Rev. 2021;24(5):723\u0026ndash;46. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/14413523.2021.1902143\u003c/span\u003e\u003cspan address=\"10.1080/14413523.2021.1902143\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOlden A, M\u0026oslash;en J. The triple difference estimator. Econometrics J. 2022;25(3):531\u0026ndash;53. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/ectj/utac010\u003c/span\u003e\u003cspan address=\"10.1093/ectj/utac010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOlvera Alvarez HA, Appleton AA, Fuller CH, Belcourt A, Kubzansky LD. An Integrated Socio-Environmental Model of Health and Well-Being: A Conceptual Framework Exploring the Joint Contribution of Environmental and Social Exposures to Health and Disease Over the Life Span. Curr Environ Health Rep. 2018;5(2):233\u0026ndash;43. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s40572-018-0191-2\u003c/span\u003e\u003cspan address=\"10.1007/s40572-018-0191-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark AH, Zhong S, Yang H, Jeong J, Lee C (n.d.), editors. Impact of COVID-19 on physical activity: A rapid review. \u003cem\u003eJournal of Global Health\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e, 05003. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7189/jogh.12.05003\u003c/span\u003e\u003cspan address=\"10.7189/jogh.12.05003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePowell LM, Slater S, Chaloupka FJ, Harper D. Availability of Physical Activity\u0026ndash;Related Facilities and Neighborhood Demographic and Socioeconomic Characteristics: A National Study. Am J Public Health. 2006;96(9):1676\u0026ndash;80. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2105/AJPH.2005.065573\u003c/span\u003e\u003cspan address=\"10.2105/AJPH.2005.065573\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReijneveld SA, Verheij RA, de Bakker DH. The impact of area deprivation on differences in health: Does the choice of the geographical classification matter? J Epidemiol Commun Health. 2000;54(4):306\u0026ndash;13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/jech.54.4.306\u003c/span\u003e\u003cspan address=\"10.1136/jech.54.4.306\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRivera-Navarro J, Bonilla L, Gullon P, Gonzalez-Salgado I, Franco M. Can we improve our neighbourhoods to be more physically active? Residents\u0026rsquo; perceptions from a qualitative urban health inequalities study. Health Place. 2022;77:102658. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.healthplace.2021.102658\u003c/span\u003e\u003cspan address=\"10.1016/j.healthplace.2021.102658\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRocha V, Ribeiro AI, Severo M, Barros H, Fraga S. Neighbourhood socioeconomic deprivation and health-related quality of life: A multilevel analysis. PLoS ONE. 2017;12(12):e0188736. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0188736\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0188736\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRogers AE, Schenkelberg MA, Stoepker P, Westmark D, Srivastava D, Dzewaltowski DA. Indicators of community physical activity resources and opportunities and variation by community sociodemographic characteristics: A scoping review. Prev Med Rep. 2024;40:102656. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.pmedr.2024.102656\u003c/span\u003e\u003cspan address=\"10.1016/j.pmedr.2024.102656\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSalvo G, Lashewicz BM, Doyle-Baker PK, McCormack GR. Neighbourhood Built Environment Influences on Physical Activity among Adults: A Systematized Review of Qualitative Evidence. Int J Environ Res Public Health. 2018;15(5):897. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijerph15050897\u003c/span\u003e\u003cspan address=\"10.3390/ijerph15050897\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSantacruz Lozano JA, Espada Mateos M, Clemente Remon A, Navarro J-B. J. E. (2021). Spanish sport facilities: Differences between public and private, and according to their business model. \u003cem\u003eRETOS-NUEVAS TENDENCIAS EN EDUCACION FISICA DEPORTE Y RECREACION\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e, 38\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSawyer A, Ucci M, Jones R, Smith L, Fisher A. Simultaneous evaluation of physical and social environmental correlates of physical activity in adults: A systematic review. SSM-POPULATION HEALTH. 2017;3:506\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ssmph.2017.05.008\u003c/span\u003e\u003cspan address=\"10.1016/j.ssmph.2017.05.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSFM C, Van Cauwenberg J, Maenhout L, Cardon G, Lambert EV, Van Dyck D. Inequality in physical activity, global trends by income inequality and gender in adults. Int J Behav Nutr Phys Activity. 2020;17(1):142. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12966-020-01039-x\u003c/span\u003e\u003cspan address=\"10.1186/s12966-020-01039-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShen J, Cheng J, Huang W, Zeng F. An Exploration of Spatial and Social Inequalities of Urban Sports Facilities in Nanning City. China Sustain. 2020;12(11):4353. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su12114353\u003c/span\u003e\u003cspan address=\"10.3390/su12114353\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSinger JD, Willett JB. (2003). \u003cem\u003eApplied longitudinal data analysis: Modeling change and event occurrence\u003c/em\u003e (pp. xx, 644). Oxford University Press. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/acprof:oso/9780195152968.001.0001\u003c/span\u003e\u003cspan address=\"10.1093/acprof:oso/9780195152968.001.0001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStataCorp. (2021). \u003cem\u003eStata Statistical Software: Release 17.College Station, TX: StataCorp LLC.\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://stata-nz.com/docs/manuals/stata17/p.pdf\u003c/span\u003e\u003cspan address=\"https://stata-nz.com/docs/manuals/stata17/p.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStokols D. (1996). Translating social ecological theory into guidelines for community health promotion. \u003cem\u003eAMERICAN JOURNAL OF HEALTH PROMOTION\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(4), 282\u0026ndash;298. Art and Science of Health Promotion Conference. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4278/0890-1171-10.4.282\u003c/span\u003e\u003cspan address=\"10.4278/0890-1171-10.4.282\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun P, Sun J, Jin L, Zhu Y. COVID-19 pandemic changes the outdoor physical activity preference in Chinese city: A 7-year GPS trajectory data analysis. Cities. 2024;152:105253. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cities.2024.105253\u003c/span\u003e\u003cspan address=\"10.1016/j.cities.2024.105253\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Lenthe FJ, Brug J, Mackenbach JP. Neighbourhood inequalities in physical inactivity: The role of neighbourhood attractiveness, proximity to local facilities and safety in the Netherlands. Soc Sci Med. 2005;60(4):763\u0026ndash;75. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.socscimed.2004.06.013\u003c/span\u003e\u003cspan address=\"10.1016/j.socscimed.2004.06.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Y, Xie H, Hua Y, Jin B, Qiu Z. Theoretical framework of life circles in Chinese small towns and the optimization of spatial layout for public service facilities based on residents\u0026rsquo; distance sensitivity. Humanit Social Sci Commun. 2024;11(1):1\u0026ndash;14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1057/s41599-024-04321-6\u003c/span\u003e\u003cspan address=\"10.1057/s41599-024-04321-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhite RL, Vella S, Biddle S, Sutcliffe J, Guagliano JM, Uddin R, Burgin A, Apostolopoulos M, Nguyen T, Young C, Taylor N, Lilley S, Teychenne M. Physical activity and mental health: A systematic review and best-evidence synthesis of mediation and moderation studies. Int J Behav Nutr Phys ACTIVITY. 2024;21(1):134. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12966-024-01676-6\u003c/span\u003e\u003cspan address=\"10.1186/s12966-024-01676-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWHO. (2020). \u003cem\u003eWHO guidelines on physical activity and sedentary behaviour\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.who.int/publications/i/item/9789240015128\u003c/span\u003e\u003cspan address=\"https://www.who.int/publications/i/item/9789240015128\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu R, Yue W, Wei F, Yang G, Chen Y, Pan K. Inequality of public facilities between urban and rural areas and its driving factors in ten cities of China. Sci Rep. 2022;12(1):13244. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-022-17569-2\u003c/span\u003e\u003cspan address=\"10.1038/s41598-022-17569-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, Liu. An age-stratified study on the accessibility to sports venues and their impact on resident satisfaction: Evidence from urban Beijing. Progress Geogr. 2024;43(7):1416\u0026ndash;28. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.18306/dlkxjz.2024.07.011\u003c/span\u003e\u003cspan address=\"10.18306/dlkxjz.2024.07.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang Y, Peng C, Yeung CY, Ren C, Luo H, Lu Y, Yip PSF, Webster C. Moderation effect of visible urban greenery on the association between neighbourhood deprivation and subjective well-being: Evidence from Hong Kong. Landsc Urban Plann. 2023;231:104660. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.landurbplan.2022.104660\u003c/span\u003e\u003cspan address=\"10.1016/j.landurbplan.2022.104660\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y, Koene M, Chen C, Wagenaar C, Reijneveld SA. Associations between the built environment and physical activity in children, adults and older people: A narrative review of reviews. Prev Med. 2024;180:107856. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ypmed.2024.107856\u003c/span\u003e\u003cspan address=\"10.1016/j.ypmed.2024.107856\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y, Ming Y, Y., Shi B. Spatial distribution characteristics and causes of public sports venues in China. Sci Rep. 2023;13(1):15056. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-023-42308-6\u003c/span\u003e\u003cspan address=\"10.1038/s41598-023-42308-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou P, Hu Z, Chen Y, Liu K, Wang Y. Parenthood, spatial temporal environmental exposure, and leisure-time physical activity participation: Evidence from a micro-timescale retrospective longitudinal study. Health Place. 2024;85:103170. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.healthplace.2023.103170\u003c/span\u003e\u003cspan address=\"10.1016/j.healthplace.2023.103170\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Sport facility accessibility, Built environment, Physical activity, Health inequity, Neighborhood deprivation, Natural experiment, Public health policy","lastPublishedDoi":"10.21203/rs.3.rs-6484311/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6484311/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eaccessibility to physical activity (PA) facility can strongly influence population health outcomes through PA engagement duration in neighborhoods. While commercial facilities tend to spatially cluster, creating inequitable PA opportunities and health disparities between neighborhoods, little is known about how public facilities buffer against such neighborhood health inequities. Regulation of facility openings during the COVID-19 pandemic provided a natural experimental setting to examine this relationship.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe conducted a longitudinal study of PA behaviors among 701 residents from 23 neighborhoods in Shenzhen during facility closure and reopening (2019\u0026ndash;2020). Using difference-in-differences (DID) analysis, we examined how different types of facilities influenced various PA duration. Through triple-difference (DDD) analysis, we investigated how public facilities moderated PA duration responses to commercial facility changes within 1,000-meter catchments.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003ePA facilities showed distinct functional effects, with parks increasing light physical activity (LPA) (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;39.45, p\u0026thinsp;=\u0026thinsp;0.032) and public sports facilities enhancing moderate-to-vigorous physical activity (MVPA) (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;39.07, p\u0026thinsp;=\u0026thinsp;0.015); Commercial facilities exhibited spatial polarization with diminishing marginal returns, showing strong effects on MVPA in low-accessibility areas (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;70.02, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) but negligible impact in high-accessibility areas ; Through triple-difference analysis, we quantified the conditional buffering effect where public sports facilities effectively mitigated MVPA reduction in areas experiencing commercial facility disadvantages (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;79.27, p\u0026thinsp;=\u0026thinsp;0.050), with this effect being strongest in commercially deprived areas (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;42.51, p\u0026thinsp;=\u0026thinsp;0.001).\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThrough spatially balanced distribution and complementary functional design, public sports facilities effectively mitigated neighborhood MVPA inequities caused by commercial facility clustering. This conditional buffering effect was particularly significant in commercially disadvantaged areas, informing public health and urban planning policies for building resilient and equitable urban PA environments.\u003c/p\u003e","manuscriptTitle":"Buffering effects of public sports facilities on physical activity and health equity during commercial facility fluctuations: A natural experiment in Shenzhen, China","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-20 12:38:28","doi":"10.21203/rs.3.rs-6484311/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7ec5c65f-4856-4038-87a8-b7fdf246fa5a","owner":[],"postedDate":"May 20th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-11T14:41:18+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-20 12:38:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6484311","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6484311","identity":"rs-6484311","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}