Abstract
Background
The optimal scale of urbanization for stroke health care provision and the potential impact of compact city design on stroke outcomes remain unclear. We investigated the impact of zip code area‐level compact city design using the walkability index (WI) and its mediators on stroke outcomes.
Methods
This nationwide retrospective study used data from patients with stroke from the J‐ASPECT study (2017–2022). WI was calculated as the average of 3 Z‐scored city design elements (population density, road connectivity, and variation in walkable facilities) from 113 1156 zip code areas in Japan. The association between WI and in‐hospital mortality, functional independence at discharge, and medical costs was assessed using multivariable mixed‐effects logistic regression model.
Results
Overall, 555 296 patients (median age, 75 [interquartile range, 66–83] years; female, 42.5%) from 818 hospitals were included. Higher WI was significantly associated with decreased in‐hospital mortality (odds ratio [OR], 0.94 [95% CI, 0.92–0.96]) and increased functional independence (OR, 1.03 [95% CI, 1.02–1.04]). The highest WI group was associated with decreased mortality, primarily mediated by management in intensive or stroke care units (proportion mediated, 0.46 [95% CI, 0.35–0.63]), and the highest WI group was associated with increased functional independence, mediated by short road distance to the hospital (proportion mediated, 0.30 [95% CI, 0.21–0.44]).
Conclusions
Zip code area‐level compact city design was associated with decreased in‐hospital mortality and increased functional independence. Compact city design at community level, even without large‐scale urbanization, may contribute to improving stroke care provision and outcomes in increasingly urbanized societies.
Keywords: city design, city environment, compact city, primary stroke center, stroke, urban, walkability
Subject Categories: Cerebrovascular Disease/Stroke, Health Equity, Social Determinants of Health, Epidemiology, Lifestyle
Nonstandard Abbreviations and Acronyms
- ADI
area deprivation index
- ICH
intracerebral hemorrhage
- IS
ischemic stroke
- PSC
primary stroke center
- WI
walkability index
Clinical Perspective.
What Is New?
This study revealed the significant association between community‐level compact city design and stroke outcomes.
The association was mediated by prehospital and hospital factors, which seemed to reflect the burden of stroke care provision.
What Are the Clinical Implications?
Compact city design at community level, even without large‐scale urbanization, may contribute to improving stroke care provision and outcomes in increasingly urbanized societies.
Urbanization is accelerating in developed countries. Currently, >50% of the world's population lives in urban areas, and this proportion may reach 70% by 2050. 1 With rapid urbanization, built environmental factors have become highly modifiable, and city design is a major concern in providing solutions to health care issues. Globally, certification of stroke centers and regional network establishment for qualified and sustainable stroke health care is increasing; however, disparities in stroke treatment access remain. 2 , 3 , 4 , 5 , 6 Having the most aged population globally, Japan faces the social issue of efficient system development from the regional urban planning perspective to reduce disparities in human and material resources. In this context, the impact of city design on stroke outcomes and its optimal scale for stroke health care provision warrants discussion.
Walkability is a representative urban planning model concept that promotes higher levels of density, diversity, destination accessibility, and intensified urban form as exemplified by the “compact city” concept. 7 Higher walkability reduces vehicle use, promotes public transport use, walking, and encourages the use of facilities within walking distance, revitalizing the local economy and residents' health. Portland, Oregon in the United States is a positive compact city example—a concept that has also been adopted in Japan since 2015 by the national and local governments for sustainable urbanization. 8 , 9 Walkability index (WI) is associated with cardiovascular events and mortality. 10 , 11 , 12 , 13 , 14 However, although the relationship between WI and stroke incidence is known, to our knowledge, this relationship between WI and stroke outcomes has not been evaluated considering the impact of health care provision.
Although population‐ or region‐based urban–rural disparities in stroke health care availability have been reported, 2 , 3 , 4 , 5 , 6 the potential impact of city design and optimal urban scale for acute stroke care provision is unknown. Thus, this study investigated the impact of compact city design and its mediators on health care availability for stroke outcomes using the zip code area‐level WI and individual patient data in Japan.
METHODS
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Data Source and Study Population
This nationwide cohort study was conducted using data from the J‐ASPECT study: a nationwide survey of the Acute Stroke Care Capacity for Proper Designation of Comprehensive Stroke Center in Japan, launched in 2010 as the first nationwide survey of acute stroke clinical practices using data from the Japanese health insurance claims database. 15 , 16 We identified all consecutive admitted patients with International Classification of Diseases, Tenth Revision (ICD‐10) code of main disease or most medical expense use being ischemic stroke (I63), intracerebral hemorrhage (ICH) (I61 excluded I619), and subarachnoid hemorrhage (SAH) (I60) from the diagnosis procedure combination claim database of 850 hospitals between April 2017 to March 2022. The inclusion criteria were emergency admission and admission within 3 days of stroke onset, excluding admission for diagnostic purposes, rehabilitation, and elective surgery. Patients with missing zip code information and those who had lived in facilities or were hospitalized just before admission were also excluded. Those admitted to a prefecture distant from their zip code (not geographically adjacent) were also excluded as tourists. The patient selection flow chart is shown in Figure 1. This study was approved by the institutional review boards of the participating centers; given its retrospective nature, the requirement for informed consent was waived. This study conformed to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.
Figure 1. Patient selection flow chart.
Emergency admission and admission within 3 days of stroke onset were included. Patients with missing zip code information, those who had lived in facilities, or those who were hospitalized just before admission were excluded. Those admitted to a prefecture distant from their zip code (not geographically adjacent) were also excluded.
Walkability Index and Other Variables
To evaluate community‐level compact urbanization in each patient's residence area, all patients were assessed using the zip code area‐level WI. WI was calculated as the average of 3 Z‐scored variables: population density, road connectivity (intersection density), and walkable destinations (number of facility types). The representative point was established as the center of the highest population density fifth grid square in the 2020 census for each zip‐code area. These 3 city design variables were calculated for an area within a 1‐km road distance from the representative point as a walkable area using a geographical information system. A higher WI indicates a more compact and intensified urbanization. The precise developmental process of this index is reported in Data S1. 17 The WI distribution of all 113 156 evaluable zip code areas in Japan is shown in Figure S1. We categorized all zip code areas into quartile groups using the WI (first group: WI<–0.80, second group: –0.80≤WI<0.04, third group: 0.04≤WI<0.76, fourth group: 0.76≤WI). Zip codes in the fourth group constituted the most compact urban, whereas those in the first group constituted the most diffuse areas (Figure 2). The geographical distribution of the WI groups, with all the representative points in Japan, is shown in Figure 3. All patients were classified into the 4 categorical WI groups according to their zip code.
Figure 2. Representative community categorized into first and fourth walkability index groups.
The variation of facility type is represented as 18 different colors. The community of the fourth WI group looks more intense with intersection dots and facility diamonds compared with the community of the first WI group. WI indicates walkability index.
Figure 3. Distribution of walkability index group and zip code area‐level representative points.
The representative points of all 113 156 zip code areas in Japan were assigned to WI and categorized into quartile groups by WI. The higher the WI area, the darker the color. WI indicates walkability index.
The following baseline characteristics were collected: age, sex, stroke risk factors (hypertension, diabetes, dyslipidemia, and smoking), region (east or west), body mass index, premodified Rankin Scale score 0–2, hospital frailty risk score, 18 stroke subtypes (SAH, ICH, or ischemic stroke), and consciousness level at admission (awake, arousable, or comatose). Zip code area‐level compact city factors (population density, number of facility types, and intersection density) and area deprivation index (ADI) were collected. ADI was calculated as zip code area‐level socioeconomic status using Census 2020 by Nakaya et al., combining the proportion of old couple households, old single households, rent houses, sales and service workers, agricultural workers, blue‐collar workers, and unemployment rate. 19 Patient information regarding stroke health care accessibility and received treatment were collected: road distance to the hospital, ambulance use, interhospital transfer, admission across secondary medical area boundary, off‐hour admission, admission at a primary stroke center (PSC) and PSC core (equivalent to a comprehensive stroke center in the United States, certified by Japanese Stroke Society), tissue‐type plasminogen activator use, mechanical thrombectomy, surgical treatment including endovascular intervention, and management in stroke care or intensive care unit.
The outcomes assessed were in‐hospital mortality, functional independence (defined as modified Rankin Scale scores ranging from 0 to 2) at discharge, and medical costs.
Statistical Analysis
Baseline characteristics, zip code area‐level factors, stroke health care accessibility, and received treatment were summarized descriptively and compared across WI groups using χ2 and Kruskal–Wallis tests. We examined the association between WI and the outcomes and analyzed the mediation of the association through stroke health care availability. For association analysis, we fitted multivariable mixed‐effects logistic regression models, with the 1886 municipalities as a random effect. Covariates were selected based on prior literature, clinical relevance, and data availability. A stepwise modeling approach was used: Model 1 was unadjusted, Model 2 included demographic and baseline functional characteristics, and Model 3 further adjusted for stroke risk factors and clinical subtype and severity. For the medical cost outcome, surgical treatment was also included (Data S2). WI was inputted as a continuous variable and a categorical variable of the 4 groups based on its quartiles (the lowest, first group acting as a reference). Odds ratios (ORs) and 95% CIs for each WI quartile were estimated using individual Wald tests. To examine linear trends across WI quartiles, we additionally modeled WI quartile as an ordinal variable and calculated the P for trend. No formal omnibus test was conducted. To systematically evaluate how the impact of WI on outcomes varied by subpopulation, we also performed subgroup analyses by age (<65 and ≥65 years), sex, stroke subtype, consciousness level, and ADI quartile (Q1–Q4).
In the mediation analysis, using a counterfactual scenario, we quantified the effects of the highest (fourth) WI group on the outcomes acting through 4 prespecified possible prehospital mediators (increased prevalence of interhospital transfer, decreased admission across medical area boundary, shorter road distance to the hospital, and increased prevalence of off‐time admission) and 4 possible hospital mediators (increased prevalence of mechanical thrombectomy, surgical treatment including endovascular intervention, management in intensive or stroke care unit, and admission in PSC core). Mediation analysis was performed using “mediation package” in R and all effects were suggested as risk difference. Details of the association and mediation analyses are provided in Data S2.
Continuous and categorical variables are presented as median (interquartile range) and frequencies (%), respectively. Statistical multiplicity was not considered. Statistical significance was set at P value <0.05. All statistical analyses were performed using R software, version 4.3.3 (R Foundation for Statistical Computing, Vienna, Austria).
RESULTS
Patient Characteristics
Overall, 555 296 patients at 818 hospitals were included from the J‐ASPECT study. Patient baseline characteristics are summarized in Table 1. Their median age was 75 (interquartile range, 66–83) years, and 42.5% were female. The distribution of all patients according to WI is shown in Figure S2. Respectively, 32 058, 90 864, 161 106, and 271 268 patients were categorized into the first (WI<−0.80), second (−0.80≤WI<0.04), third (0.04≤WI<0.76), and fourth (0.76≤WI) groups.
Table 1.
Baseline Characteristics, Stroke Information, Zip Code Area‐Level Factors, and Stroke Health Care Availability According to Walkability Index Groups
| Characteristics | Total | Walkability index groups | P value | |||
|---|---|---|---|---|---|---|
| First (WI <−0.80) | Second (−0.80≤WI<0.04) | Third (0.04≤WI<0.76) | Fourth (0.76≤WI) | |||
| No. | 555 296 | 32 058 | 90 864 | 161 106 | 271 268 | |
| Patient characteristics | ||||||
| Age, y, median (IQR) | 75 (66 to 83) | 78 (68 to 85) | 76 (67 to 84) | 75 (66 to 83) | 75 (65 to 82) | <0.001 |
| Female sex, No. (%) | 236 169 (42.5) | 14 202 (44.3) | 39 191 (43.1) | 68 182 (42.3) | 114 594 (42.2) | <0.001 |
| Hypertension, No. (%) | 311 860 (56.2) | 18 319 (57.1) | 52 117 (57.4) | 90 242 (56.0) | 151 182 (55.7) | <0.001 |
| Diabetes, No. (%) | 120 892 (21.8) | 6513 (20.3) | 19 542 (21.5) | 35 755 (22.2) | 59 082 (21.8) | <0.001 |
| Dyslipidemia, No. (%) | 154 044 (27.7) | 7569 (23.6) | 23 496 (25.9) | 44 071 (27.4) | 78 908 (29.1) | <0.001 |
| Current/past smoker, No. (%) | 168 164 (34.9) | 8832 (30.9) | 26 922 (33.3) | 49 906 (35.2) | 82 504 (35.8) | <0.001 |
| Region, No. (%) | <0.001 | |||||
| East | 314 255 (56.6) | 18 430 (57.5) | 52 627 (57.9) | 97 571 (60.6) | 145 627 (53.7) | |
| West | 241 041 (43.4) | 13 628 (42.5) | 38 237 (42.1) | 63 535 (39.4) | 125 641 (46.3) | |
| Body mass index, median (IQR), kg/m2 | 22.8 (20.4 to 25.3) | 22.7 (20.3 to 25.1) | 22.7 (20.4 to 25.2) | 22.8 (20.4 to 25.3) | 22.8 (20.4 to 25.3) | <0.001 |
| Premodified Rankin Scale 0–2, No. (%) | 467 639 (84.8) | 26 319 (82.5) | 75 609 (83.6) | 136 214 (85.0) | 229 497 (85.3) | <0.001 |
| Charlson score, median (IQR) | 5 (5 to 6) | 5 (5 to 6) | 5 (5 to 6) | 5 (4 to 6) | 5 (4 to 6) | <0.001 |
| Hospital Frailty Risk Score, No. (%) | 0.5 (0.0 to 2.6) | 0.8 (0.0 to 3.0) | 0.8 (0.0 to 2.6) | 0.5 (0.0 to 2.6) | 0.5 (0.0 to 2.6) | <0.001 |
| Stroke information | ||||||
| Stroke subtypes, No. (%) | <0.001 | |||||
| Subarachnoid hemorrhage | 38 915 (7.0) | 2256 (7.0) | 6551 (7.2) | 11 869 (7.4) | 18 239 (6.7) | |
| Intracerebral hemorrhage | 124 492 (22.4) | 7319 (22.8) | 20 636 (22.7) | 36 093 (22.4) | 60 444 (22.3) | |
| Ischemic stroke | 391 889 (70.6) | 22 483 (70.1) | 63 677 (70.1) | 113 144 (70.2) | 192 585 (71.0) | |
| Consciousness level | <0.001 | |||||
| Awake | 437 941 (78.9) | 24 342 (75.9) | 70 138 (77.2) | 126 465 (78.5) | 216 996 (80.0) | |
| Arousable | 61 219 (11.0) | 3942 (12.3) | 10 725 (11.8) | 18 193 (11.3) | 28 359 (10.5) | |
| Comatose | 56 135 (10.1) | 3774 (11.8) | 10 001 (11.0) | 16 447 (10.2) | 25 913 (9.6) | |
| Area factors | ||||||
| Walkability index, median (IQR) | 0.74 (0.15 to 1.21) | −1.06 (−1.28 to −0.92) | −0.32 (−0.54 to −0.13) | 0.47 (0.28 to 0.63) | 1.22 (0.98 to 1.51) | <0.001 |
| Population density, median (IQR) | 4494.9 (1885.4 to 9032.8) | 310.9 (208.8 to 438.3) | 921.1 (634.1 to 1298.6) | 3049.7 (2269.3 to 3923.1) | 9150.7 (6232.6 to 12951.5) | <0.001 |
| Road connectivity, median (IQR) | 170.4 (103.6 to 230.5) | 18.2 (11.5 to 28.1) | 63.9 (43.6 to 83.6) | 139.3 (115.7 to 163.8) | 231.5 (199.6 to 274.6) | <0.001 |
| Variation in facility, median (IQR) | 14 (10 to 16) | 1 (0 to 2) | 6 (4 to 9) | 12 (10 to 14) | 16 (15 to 17) | <0.001 |
| Area deprivation index, median (IQR) | −0.05 (−0.61 to 0.53) | 0.68 (0.09 to 1.35) | 0.27 (−0.23 to 0.81) | 0.03 (−0.42 to 0.54) | −0.31 (−0.88 to 0.27) | <0.001 |
| Stroke health care availability | ||||||
| Road distance to hospital, median (IQR), km | 6.2 (3.4 to 12.2) | 21.0 (13.2 to 33.4) | 13.3 (8.0 to 22.6) | 7.1 (4.0 to 12.5) | 4.2 (2.5 to 6.8) | <0.001 |
| Ambulance use, No. (%) | 370 555 (66.7) | 21 940 (68.4) | 60 271 (66.3) | 105 256 (65.3) | 183 088 (67.5) | <0.001 |
| Admission across medical boundary, No. (%) | 81 579 (14.7) | 7619 (23.8) | 18 278 (20.1) | 22 460 (13.9) | 33 222 (12.3) | <0.001 |
| Interhospital transfer, No. (%) | 36 366 (6.5) | 1782 (5.6) | 5438 (6.0) | 9608 (6.0) | 19 538 (7.2) | <0.001 |
| Off‐time admission | 234 720 (42.3) | 13 042 (40.7) | 37 700 (41.5) | 67 046 (41.6) | 116 932 (43.1) | <0.001 |
| PSC core (equal to comprehensive stroke center in the United States) | 256 084 (46.1) | 10 710 (33.4) | 32 713 (36.0) | 67 627 (42.0) | 145 034 (53.5) | <0.001 |
| PSC | 522 523 (94.1) | 29 947 (93.4) | 84 896 (93.4) | 151 577 (94.1) | 256 103 (94.4) | <0.001 |
| Tissue‐type plasminogen activator use, No. (%) | 30 587 (5.5) | 1695 (5.3) | 5026 (5.5) | 9018 (5.6) | 14 848 (5.5) | 0.102 |
| Mechanical thrombectomy, No. (%) | 23 104 (4.2) | 1297 (4.0) | 3601 (4.0) | 6367 (4.0) | 11 839 (4.4) | <0.001 |
| Surgical treatment (including endovascular treatment) | 97 122 (17.5) | 5181 (16.2) | 15 278 (16.8) | 27 711 (17.2) | 48 952 (18.0) | <0.001 |
| Management in care unit | 153 957 (27.7) | 4539 (14.2) | 15 130 (16.7) | 34 169 (21.2) | 100 119 (36.9) | <0.001 |
IQR indicates interquartile range; PSC, primary stroke center; and WI, walkability index.
Compared with the lowest (first) WI group, the highest (fourth) had younger patients (78 [68–85] years versus 75 [65–82] years) and fewer women (44.3% versus 42.2%). The highest WI group had a lower prevalence of hypertension, higher prevalence of dyslipidemia, smoking, and premodified Rankin Scale score of 0 to 2 than the other groups. Consciousness level at admission was more severe in the lowest WI group. Regarding zip code area‐level factors and stroke health care availability, patients in the highest WI group had a lower ADI (ie, better socioeconomic status) and tended to be closer (shorter road distance) to the hospital, more often underwent interhospital transfer, were admitted across medical boundaries and in PSC and PSC cores, received surgical treatment—including endovascular intervention, and were managed in the care unit.
Association Analyses
The results of the association analysis are shown in Figure 4. Higher WI group was significantly associated with a lower proportion of in‐hospital mortality, dose‐dependently (P for trend <0.001). Compared with the lowest WI group, the highest WI group was significantly associated with decreased in‐hospital mortality (3421/32 058 (10.7%) versus 20 802/271 268 (7.7%), Model 3: OR, 0.93 [95% CI, 0.87–0.99]). Higher WI as a continuous variable was also significantly associated with decreased in‐hospital mortality (Model 3: OR, 0.94 [95% CI, 0.92–0.96]).
Figure 4. Odds ratio or percent change of the outcomes according to walkability index.
Model 1: Unadjusted using only municipalities as a random effect. Model 2: adjusted for age, female sex, premodified Rankin Scale score of 0 to 2, area deprivation index. Model 3 when in‐hospital mortality or functional independence was the outcome: adjusted for age, female sex, hypertension, diabetes, dyslipidemia, current/past smoker, stroke subtypes, consciousness level, premodified Rankin Scale score of 0 to 2, area deprivation index. Model 3 when medical costs was the outcome: adjusted for age, female sex, hypertension, diabetes, dyslipidemia, current/past smoker, stroke subtypes, consciousness level, premodified Rankin Scale score of 0 to 2, area deprivation index, and surgical and endovascular intervention. A, Analysis of in‐hospital mortality. B, Analysis of functional independence at discharge. C, Analysis of medical costs. Model 3 forest plots are shown on the right side. IQR indicates interquartile range; JPY, Japanese yen; mRS, modified Rankin Scale; OR, odds ratio; and WI, walkability index.
Higher WI group was significantly associated with a higher proportion of patients with functional independence at discharge dose‐dependently (P for trend <0.001). Compared with the lowest WI group, the highest was significantly associated with increased functional independence at discharge (13 945/32 058 (43.6%) versus 141 156/271 268 (52.4%), Model 3: OR, 1.05 [95% CI, 1.02–1.09]). Higher WI as a continuous variable was also significantly associated with increased functional independence at discharge (Model 3: OR, [1.03 [95% CI, 1.02–1.04]).
There were no significant associations between WI group and medical costs (P for trend <0.803), nor between medical cost and the highest WI group compared with the lowest (median, 1.08 [interquartile range, 0.64–1.85] million Japanese yen versus 1.08 [interquartile range, 0.64–1.90] million Japanese yen, Model 3: percent change, −0.55 [95% CI, −1.33 to 0.79]) or higher WI as a continuous variable (Model 3: percent change, −0.12 [95% CI, −0.52 to 0.27]).
Three city design factors constituting WI (population density, intersection density, and number of facility types) were independently confirmed to be significantly associated with decreased in‐hospital mortality and increased functional independence at discharge (Table S1).
Subgroup Analyses
The results of the subgroup analyses are shown in Table 2. Across all subgroups, higher WI was significantly associated with decreased in‐hospital mortality and increased functional independence at discharge. Notably, patients younger than 65 years (Model 3: OR, 0.91 [95% CI, 0.87–0.96]), male (Model 3: OR, 0.92 [95% CI, 0.90–0.95]), those with SAH (Model 3: OR, 0.90 [95% CI, 0.86–0.95]), those who were comatose (Model 3: OR, 0.92 [95% CI, 0.89–0.95]), and individuals in the first ADI quartile (highest economic status) group (Model 3: OR, 0.90 [95% CI, 0.86–0.94]) demonstrated the lowest ORs for the association between WI and in‐hospital mortality within their respective demographic groups. Patients younger than 65 years (Model 3: OR, 1.05 [95% CI, 1.02–1.08]), female (Model 3: OR, 1.07 [95% CI, 1.04–1.09]), those with ICH (Model 3: OR, 1.06 [95% CI, 1.03–1.09]), those who were comatose (Model 3: OR, 1.14 [95% CI, 1.08–1.21]), and individuals in the first ADI quartile (highest economic status) group (Model 3: OR, 1.05 [95% CI, 1.01–1.08]) demonstrated the highest ORs for the association between WI and functional independence within demographic categories. Patients younger than 65 years (Model 3: percent change, 1.13 [95% CI, 0.33–1.92]), those with SAH (Model 3: percent change, 3.79 [95% CI, 2.63–4.95]) and ICH (Model 3: percent change, 0.84 [95% CI, 0.00–1.69]), and those who were arousable (Model 3: OR, 2.00 [95% CI, 1.07–2.94]) and comatose (Model 3: OR, 4.30 [95% CI, 2.97–5.64]) demonstrated significant association between higher WI and increased medical cost.
Table 2.
Odds Ratio and Percent Change of the Outcomes According to Walkability Index in Specific Subgroups
| Subgroups | In‐hospital mortality | Functional independence at discharge | Medical cost | |||
|---|---|---|---|---|---|---|
| Model 3, OR (95% CI) | P value for interaction | Model 3, OR (95% CI) | P value for interaction | Model 3, Percent change (95% CI) | P value for interaction | |
| Age | 0.671 | <0.001 | 0.173 | |||
| <65 y | 0.91 (0.87 to 0.96) | 1.05 (1.02 to 1.08) | 1.13 (0.33 to 1.92) | |||
| ≥65 y | 0.94 (0.92 to 0.96) | 1.04 (1.02 to 1.06) | −0.05 (−0.51 to 0.41) | |||
| Sex | 0.871 | <0.001 | 0.833 | |||
| Male | 0.92 (0.90 to 0.95) | 1.03 (1.02 to 1.05) | 0.46 (−0.08 to 0.99) | |||
| Female | 0.93 (0.90 to 0.96) | 1.07 (1.04 to 1.09) | −0.07 (−0.66 to 0.53) | |||
| Stroke subtypes | 0.345 | 0.052 | <0.001 | |||
| Subarachnoid hemorrhage | 0.90 (0.86 to 0.95) | 1.04 (1.00 to 1.09) | 3.79 (2.63 to 4.95) | |||
| Intracerebral hemorrhage | 0.94 (0.91 to 0.97) | 1.06 (1.03 to 1.09) | 0.84 (0.00 to 1.69) | |||
| Ischemic stroke | 0.93 (0.90 to 0.96) | 1.03 (1.02 to 1.05) | −0.04 (−0.51 to 0.42) | |||
| Consciousness level | 0.40 | <0.001 | <0.001 | |||
| Awake | 0.94 (0.91 to 0.97) | 1.03 (1.01 to 1.04) | −0.36 (−0.80 to 0.08) | |||
| Arousable | 0.92 (0.88 to 0.96) | 1.09 (1.05 to 1.13) | 2.00 (1.07 to 2.94) | |||
| Comatose | 0.92 (0.89 to 0.95) | 1.14 (1.08 to 1.21) | 4.30 (2.97 to 5.64) | |||
| Area deprivation index (quartile) | 0.200 | 0.528 | 0.781 | |||
| Q1 (highest socioeconomic status) | 0.90 (0.86 to 0.94) | 1.05 (1.01 to 1.08) | 0.45 (−0.48 to 1.38) | |||
| Q2 | 0.90 (0.86 to 0.94) | 1.04 (1.01 to 1.07) | 0.43 (−0.39 to 1.25) | |||
| Q3 | 0.92 (0.88 to 0.95) | 1.04 (1.01 to 1.06) | 0.20 (−0.58 to 0.97) | |||
| Q4 (lowest socioeconomic status) | 0.96 (0.92 to 0.99) | 1.05 (1.01 to 1.08) | 0.20 (−0.52 to 0.91) | |||
Model 3 was fitted in these subgroup analyses. Model 3 when in‐hospital mortality or functional independence was the outcome: adjusted for age, female sex, hypertension, diabetes, dyslipidemia, current/past smoker, stroke subtypes, consciousness level, pre‐mRS score of 0 to 2, area deprivation index.
Model 3 when medical costs was the outcome: adjusted for age, female sex, hypertension, diabetes, dyslipidemia, current/past smoker, stroke subtypes, consciousness level, pre‐mRS score of 0 to –2, area deprivation index, and surgical or endovascular intervention. mRS indicates modified Rankin Scale; and OR, odds ratio.
Mediation Analyses
The results of the mediation analyses are shown in Figure 5. Mediation analyses showed that the association of the highest WI group with decreased in‐hospital mortality was mediated by increased surgical or endovascular treatment (mediation effect of risk difference [RD], −0.15 [95% CI, −0.18 to −0.12], direct effect of RD, −0.94 [95% CI, −1.25 to −0.64], proportion mediated, 0.13 [95% CI, 0.09–0.19]), management in care unit (mediation effect of RD, −0.50 [95% CI, −0.53 to −0.46], direct effect of RD, [−0.59 [95% CI, −0.90 to −0.30], proportion mediated, 0.46 [95% CI, 0.35–0.63]), and admission in PSC core (mediation effect of RD, 0.13 [95% CI, 0.09–0.19], direct effect of RD, −0.95 [95% CI, −1.27 to −0.64], and proportion mediated, 0.13 [95% CI, 0.09–0.19]) (Figure 5A). Higher WI was also mediated in increased functional independence by shortened road distance to the hospital (mediation effect of RD, 0.61 [95% CI, 0.45–0.76], direct effect of RD, 1.43 [95% CI, 0.85 to −2.00], and proportion mediated, 0.30 [95% CI, 0.21–0.44]) (Figure 5B).
Figure 5. Mediation analyses for the association of the highest WI group with in‐hospital mortality and functional independence at discharge.
Adjusted for age, female sex, hypertension, diabetes, dyslipidemia, current/past smoker, stroke subtypes, consciousness level, premodified Rankin Scale score of 0 to 2, area deprivation index, the second WI group, and the third WI group. A, Mediation analyses for the association of the highest WI group with in‐hospital mortality compared with the lowest WI group in the quartile mediated by prehospital and hospital factors. B, Mediation analyses for the association of the highest WI group with functional independence at discharge compared with the lowest WI group in the quartile mediated by prehospital and hospital factors. A negative mediated proportion was considered as no mediation, as it was out of the threshold to determine a mediated proportion. int‐HP transfer indicates interhospital transfer; MT, mechanical thrombectomy; and PSC, primary stroke center.
DISCUSSION
This nationwide study revealed that zip code area‐level compact city design with a higher WI was associated with a lower prevalence of in‐hospital mortality and a higher prevalence of functional independence after stroke. These results were robust even after adjusting for patient background, areal socioeconomic status, and stroke severity and subtypes. There was no significant association between higher WI and increased medical costs, except in specific subpopulations: patients younger than 65 years, those with SAH or ICH, and those who were arousable or comatose. The significant effects of higher WI were mediated by hospital factors of increased prevalence of management in the care unit, admission to certified core stroke treatment facilities, and surgical or endovascular treatment for mortality and mediated by prehospital factors of shorter distance to the hospital for functional independence.
Although studies have reported the association between city or built environment measures and cardiovascular mortality, 20 , 21 , 22 , 23 only a few studies regarding city environment aspects such as air pollution and green space have been conducted for stroke mortality or functional outcomes globally. 24 , 25 , 26 This study found that each of the 3 city design factors (population density, road connectivity, and variation in facilities) was independently associated with stroke outcomes, suggesting that WI as an established combined index of these factors is useful to assess the impact of compact city design on stroke outcomes. Population‐ or region‐based urban–rural disparities in access to acute ischemic stroke care have also been reported. 2 , 3 , 4 , 5 , 6 Unlike these studies, our study focused on the “compact” city design relative to seeking a minimalized urban scale for acute stroke care provision. Zip code area‐level assessment enabled to perform a stratified analysis using municipalities as a mixed effect. Thus, our results seem to emphasize that community‐level intensification may improve stroke outcomes, regardless of scale, location, demographic structure, or economic profile of municipalities. Our study also included patients with both ischemic and hemorrhagic stroke, and the significant impacts of compact city design on the outcomes were confirmed in subpopulations of SAH and ICH, respectively.
From an economic perspective, compact city design can be cost effective because of the reduction in infrastructure costs and increased productivity through agglomeration economies; however, these benefits can result in higher living costs for residents. 27 , 28 Although the associations between higher WI and decreased mortality and increased functional independence were similarly demonstrated even in subgroups of individuals living in lower socioeconomic areas, a compact city design is not necessarily a panacea for people of low socioeconomic status: patients in higher WI groups tended to show relatively lower median ADI (higher socioeconomic status) in our study. Regarding medical costs, no significant relationship was confirmed between WI and stroke treatment costs in our study. This may be due to the canceling out of the effects of longer hospitalization in diffuse areas and active health care in compact cities. However, we need to recognize that compact city design may require slightly higher medical costs to improve patient outcomes, particularly in patients younger than 65 years, those with SAH or ICH, and those who are arousable or comatose.
Mediation analyses suggest the possibility that community‐level compact city design reduces mortality by providing qualified stroke care and intervention. Prompt implementation of intensive care and access to critical care reduce stroke mortality. 29 Qualified stroke health care requires increased access to stroke expertise, potential off‐hour treatment, medical imaging facilities, and an increased number of staff and care units. These issues could be resolved and a unified stroke health care system can be achieved by community‐level intensified city design. Additionally, compact city design improves functional outcomes by reducing the distance from the hospital. Remarkably, neither interhospital transfer nor mechanical thrombectomy mediated the effect of compact city design on functional outcomes. This may reflect the fact that the mechanical thrombectomy provision system has been standardized through recent stroke center certification and network formation even in diffuse areas with lower WI. Although the availability of this essential treatment is increasing, further improvements in functional outcomes can be expected by reducing the distance to hospitals through community‐level compact city planning.
Seemingly, there are other instances where a compact city design affects stroke outcomes, such as risk factor control and stroke onset awareness. Cardiovascular risk factors such as hypertension, diabetes, and obesity have attracted attention as modifiable by living in high‐walkability areas. 11 , 21 Many studies reported a higher stroke incidence in rural areas. 30 , 31 , 32 , 33 A decrease in comorbidities improves both stroke incidence and outcomes. In our cohort, fewer patients with premorbid hypertension were confirmed in the higher‐WI group. Premorbid hypertension has worse outcomes in all stroke subtypes because of diminished autoregulatory capacity and localized increase in blood–brain barrier permeability. 34 , 35 , 36 , 37 Although the effect of risk factors on stroke incidence could not be assessed in our cohort, decreased premorbid hypertension may help improve stroke outcomes in the compact city. Time delay in stroke onset awareness is also important relative to its outcomes: a delay worsens severity, increases the prevalence of comorbidities at admission, and prevents intervention opportunities. 3 , 38 , 39 , 40 A remarkable finding of this study is the low prevalence of off‐time admissions in the lower WI group. Furthermore, the prevalence of patients comatose at admission was high in the lower WI group. These results may suggest that the prevalence of stroke awareness delay is higher in the lower WI group. Social isolation in rural or diffuse areas is a major concern in an aged society. Compact cities offer the great benefits of a sense of community belonging and social interaction, 41 which may also help to improve stroke outcomes.
The risk of stroke incidence may increase owing to the intensified built environment and population density 42 , 43 ; however, evidence suggests that the risk is reduced by better access to medical institutions and a variety of facilities in more urbanized areas. 44 Although the association between compact city design and stroke incidence should be considered, the concept of community‐level compact city might aid the provision of a qualified and sustainable stroke health care system, resulting in improved outcomes.
This study has several strengths. First, it was based on the largest nationwide real‐world data set collected from stroke centers in Japan. 15 As Japan has the most aged population globally and its government is promoting urbanization based on the concept of a compact city, 8 , 9 the results of this study may offer valuable insights into the future of stroke health care in urbanized and aging societies worldwide. Second, we assessed all 555 296 individual patients using the zip code area‐level index for neighborhood walkability and ADI, in addition to having used precise information regarding patient background, stroke type and severity, and treatment. This is the first report to assess the impact of compact city design on stroke outcomes based on detailed data on stroke health care availability on a large scale in Japan. Nevertheless, this study has some limitations. First, the cohort included only Japanese patients; thus, the results may not apply to other nations. Japan achieved almost universal health care coverage; consequently, it may be relatively uncommon for individuals to refrain from seeking medical care because of financial reasons. Such background needs to be carefully considered; however, in developed countries where similar aging trends as those of Japan are expected, community‐level compact urbanization may help organize the stroke health care provision system. Second, this was a retrospective cohort study focused on patients with stroke. We could not assess the impact of a compact city design on stroke incidence rate in a comprehensive general cohort. A prospective study is needed to clarify the risk–benefit of a compact city design considering the incidence rate in the general cohort. Third, various city design factors other than those in our study can influence stroke outcomes; in our study, we selected the WI, combining only 3 factors: population density, road connectivity, and variation in facilities. These factors do not fully capture all aspects of a compact city design that could influence stroke outcomes; other factors such as access to hospitals and public transportation may also contribute. These specific indicators regarding stroke health care were assessed in the mediation analyses in this study. Future research incorporating stroke health care provision‐related indicators of compact cities would be valuable in further elucidating these relationships. Fourth, there may have been selection bias in the participating hospitals. 16 , 45 , 46
CONCLUSIONS
This largest nationwide cohort study demonstrated that zip code area‐level compact city design, evaluated using the WI, was significantly associated with reduced in‐hospital mortality and improved functional independence at discharge. These effects were consistently observed across specific subpopulations and were mediated by increased management in specialized care units and certified stroke centers, increased surgical or endovascular intervention for mortality, and a short distance to the hospital for functional independence. Community‐level compact city design may offer a potential solution for sustainable and high‐quality stroke care provision in future urbanized societies, with the expectation of improved stroke outcomes, even if it is not large‐scale urbanization.
Sources of Funding
This work was supported by the Practical Research Project for lifestyle‐related diseases, including cardiovascular diseases and diabetes, managed by the Japan Agency for Medical Research and Development (JP19ek0210088, JP20ek0210129, JP20ek0210147, JP21ek0210147, JP22ek0210147); Grants‐in‐Aid from the Japanese Ministry of Health, Labour and Welfare (H28‐Shinkin‐Ippan‐011, 19AC1003, 21FA1010, 22FA1015, 23FA1014, 24FA1015, 24FA1016); Grants‐in‐Aid for Scientific Research (KAKENHI) (25 293 314, 18H02914, 22H03191, 23K24450; principal investigator: Koji Iihara) from the Japan Society for the Promotion of Science; and Intramural Research Fund (20‐4‐10) for Cardiovascular Diseases of National Cerebral and Cardiovascular Center.
Disclosures
None.
Supporting information
J‐ASPECT Study Collaborators
Data S1
Supplemental Methods
Table S1
Figures S1–S2
Acknowledgments
The J‐ASPECT study was conducted in collaboration with the Japan Neurosurgical Society and Japan Stroke Society. We would like to thank the people and organizations that participated in the J‐ASPECT study. We would also like to thank Dr Nakaya for providing us with the zip code area‐level ADI in 2020.
A complete list of the J‐ASPECT Study Collaborators can be found in the Supplemental Material.
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.125.041293
For Sources of Funding and Disclosures, see page 12.
Contributor Information
Yukihiro Imaoka, Email: yukihiro.imaoka@gmail.com.
Koji Iihara, Email: kiihara@ncvc.go.jp.
J‐ASPECT Study Collaborators:
Takashi Shimizu, Isao Sasaki, Junta Moroi, Satoshi Okawa, Hiroaki Shimizu, Minoru Asahi, Makoto Goda, Tsuyoshi Mtsumoto, Takahisa Kano, Yoichi Katayama, Takamasa Mizuno, Shigeru Oya, Manabu Kinoshita, Takizawa Katsumi, Takao Ooasa, Taku Sato, Yukihide Kanemoto, Ryuunosuke Uranishi, Yuka Terasawa, Tetsuo Hara, Motoki Sato, Toshihiro Yamauchi, Ken Kado, Kenji Wakui, Atsushi Fujikawa, Junichiro Kumai, Okada Takashi, Satoru Hayashi, Tatuya Sihngaki, Shinji Noda, Takeshi Torigai, Takashi Tsujiuchi, Takayuki Kato, Mitsutoshi Nakada, Toshio Machida, Tomonori Kobayashi, Satoshi Fujiwara, Masahiko Tagawa, Kazuhiro Hashizume, Yujiro Tanaka, Sato Masahiro, Ichiro Nakahara, Syogo Imae, Toshiyuki Ohtani, Kei Owada, Masaki Nishimura, Yasushi Takabatake, Hiroshi Aikawa, Masaharu Tani, Kin Sigenari, Toshio Higashi, Hiroshi Abe, Yasuharu Takeuchi, Masazumi Fujii, Hiromichi Naito, Kazuo Koide, Tetsuya Tanigawara, Toru Iwama, Yoshiharu Oki, Kawamura Tsuyoshi, Satoshi Inoha, Takaaki Yamazaki, Shuji Hayashi, Hiroaki Neki, Shinji Shimato, Tadahisa Shono, Hirotaka Watarai, Yuji Nojima, Katsuzo Kiya, Shinji Obayashi, Takatoshi Fujimoto, Tadafumi Isono, Susumu Fushimi, Atsushi Saito, Masayuki Sumida, Takahito Okazaki, Taketo Kataoka, Toshiya Osanai, Akihiro Tsukada, Kazumi Nitta, Toru Kobayashi, Takayuki Sakaki, Keishi Fujita, Koichi Ishimaru, Sumio Kobayashi, Takahara Kenta, Kohsho Fujikawa, Naoaki Kanda, Hirofumi Goto, Kazuma Kowata, Katsumi Matsumoto, Naoyuki Uchiyama, Tsuyoshi Inoue, Yoshitomo Uchiyama, Shoko Atsuchi, Noda Shinya, Toshiyuki Tsuboi, Kuniaki Ogasawara, Kazuyuki Miura, Taro Suzuki, Hisashi Kubota, Tomohisa Okada, Masaru Abiko, Yosihisa Kawano, Kenichiro Fujishiro, Tsuyoshi Nakajima, Kazunari Koga, Naoyuki Imamoto, Hirofumi Hiyama, Yukio Seki, Shunsuke Shibao, Takeshi Matsuoka, Jiro Kitayama, Tsuyoshi Ichikawa, Toshinari Misaki, Keisuke Ito, Fumitaka Miya, Hidenori Yokota, Tadashi Terasaki, Keisuke Imai, Ken Asakura, Hiroki Fukuda, Tsukasa Wada, Michio Nakamura, Hideki Arakawa, Noboru Imai, Kazuya Takahashi, Kunihiko Kodama, Shogo Ogita, Keigo Matsumoto, Ryota Tanaka, Kazuo Kitazawa, Takuji Yamamoto, Kazuo Yamashiro, Naoki Shirasaki, Masanobu Okauchi, Hiroshi Tokimura, Shunichi Tanaka, Koichi Okiya, Keiji Kidoguchi, Tetsuo Ando, Nobutaka Yamamoto, Kazuyuki Nishigaya, Masayuki Mizobuchi, Morio Takasaki, Kunikazu Yoshimura, Toshihisa Nishizawa, Katsuhiko Maruichi, Kazutami Nakao, Junya Hayashi, Hiroshi Hayashi, Masayuki Sasou, Naoaki Sato, Hideaki Ishihara, Michihiro Hayasaka, Ryuhei Kouno, Satoshi Magarisawa, Osamu Kawakami, Hiroyuki Matumoto, Shigeru Miyake, Masaru Idei, Tsukamoto Haruhisa, Ryo Miyaoka, Teruo Kimura, Kenta Kunimoto, Toshihiro Kumabe, Nobuyuki Sakai, Noriaki Ashida, Takashi Tominaga, Atsushi Fujita, Hiroyuki Nishimura, Hitoshi Fukuda, Toyoaki Shinohara, Yukako Yazawa, Yasuhiko Motooka, Kenjiro Fujiwara, Taketo Hatano, Takenori Kato, Yuuichi Hirota, Akihiko Kaga, Masayuki Arai, Takuya Kawai, Yasuyuki Kaku, Masaki Chin, Hiromichi Yamamoto, Kosuke Katayama, Takeshi Ogura, Motohiro Morioka, Daisuke Shimbo, Hiroshi Sakaida, Masayuki Yokota, Tatsuhito Yamagami, Tomoyuki Ohara, Mamoru Murakami, Norio Nakajima, Norikazu Kurokawa, Seiji Gotoh, Sei Haga, Koji Yoshimoto, Yoshinori Arai, Toshiyuki Tsukada, Shigeyoshi Kimura, Mamoru Ota, Matsui Ryukichi, Kenji Ooyama, Yuji Yamamoto, Yukihiko Kawamoto, Katsuhiko Hayasi, Toshikazu Takeshima, Yotaro Takeuchi, Arai Motohiro, Fujimaro Ishida, Hidenori Suzuki, Takahiro Murata, Yoichi Harada, Takashi Matsuoka, Megumi Nitta, Osamu Hamasaki, Masanori Morimoto, Yoshimasa Kinoshita, Yosikazu Arai, Akihiro Itoh, Hiroshi Ooyama, Yasuaki Inoue, Satoshi Horiguchi, Yosikazu Kusano, Makoto Hirose, Tsuyoshi Izumo, Amano Takayuki, Hiromichi Koga, Hidemori Tokunaga, Kenta Fujimoto, Akiko Arakawa, Masatoshi Koga, Yu Takeda, Yoroyoshi Kimura, Takashi Sadatomo, Kotaro Ogihara, Katsuhiro Yamashita, Takashi Nakagawa, Shunichi Fukuda, Yasuhiro Manabe, Naoki Tokumitsu, Norihiro Ishii, Masayuki Ezura, Hideki Sakai, Miho Higashi, Atsuo Yoshino, Ken‐Ichi Morita, Kiyoshi Onda, Shunichi Yoneda, Shigeta Moriya, Masaki Sakamoto, Masao Tsuji, Shigenori Katayama, Yasuhiro Aida, Norihiko Akao, Haruki Takahashi, Masafumi Ohtaki, Naoya Shimada, Takahiro Maeda, Yoshihiro Kuga, Masahisa Kawakami, Yasuyuki Nagai, Keita Suzuki, Koji Tokunaga, Akira Handa, Isao Date, Koji Idomari, Tomoaki Nagamine, Shinobu Adachi, Toshiki Aoki, Masayuki Nakajima, Terukazu Kuramoto, Nishihara Jun, Tomoya Ishiguro, Manabu Sakaguchi, Akatsuki Wakayama, Naohiro Osaka, Hiroyuki Hashimoto, Masahiro Yoshida, Kenji Hashimoto, Kenichi Matsumoto, Yasunobu Goto, Tatunori Kawai, Toshiro Katsuta, Seisuke Iseki, Toshiro Yonehara, Hiroto Murata, Michiya Kubo, Masashi Nakatsukasa, Yuuzi Okamoto, Sunao Takemura, Makoto Inaba, Atsuhiro Kojima, Soichi Oya, Shinya Kohyama, Yuichiro Kikkawa, Toshie Takahashi, Takashi Tsuzuki, Takaaki Yoshida, Tsunenori Ozawa, Youhei Kudoh, Takigami Masayoshi, Nobuhiro Mikuni, Ken Takahashi, Tanikawa Rokuya, Kouichirou Takemoto, Shigehiro Nakahara, Kazunari Homma, Akiyoshi Sato, Soichi Akamine, Watabe Noriaki, Kazuhiko Nozaki, Shinya Okita, Takashi Yoshida, Tsutomu Hitotsumatsu, Jinichi Sasanuma, Yoshiki Hanaoka, Takashi Mizowaki, Makoto Ichinose, Shingo Yamashita, Hiroshi Shimano, Seiji Fukazawa, Shinsuke Muraoka, Masaru Honda, Yasuhiro Fujii, Kenji Kamiyama, Masahiko Kitano, Kentaro Shimoda, Noriaki Matsubara, Takashi Inoue, Zenitirou Watanabe, Shinjitsu Nishimura, Yoshihisa Fukushima, Eiji Imamura, Naomichi Wada, Shigetoshi Shimizu, Tomohiro Araki, Hiroshi Abe, Tetuya Morimoto, Kenji Shono, Masahiro Kagawa, Kazuhiko Nishino, Nobutake Sadamasa, Shoichi Shiraishi, Soichiro Komasaku, Masaaki Shojima, Katsuyuki Asaoka, Tohru Sano, Hiroki Hongo, Hidetoshi Nakamoto, Takehisa Tsuji, Morito Hayashi, Tatsuya Sasaki, Jun Yamada, Hideomi Kitajima, Hiroshi Miyazaki, Akihito Hashiguchi, Hiromi Ueta, Masahito Agawa, Yasushi Takagi, Harada Kunihiko, Kunisasu Saigusa, Masateru Katayama, Taketoshi Maehara, Kensaku Yoshida, Koichi Kato, Haruhiko Hoshino, Tatsuya Ishikawa, Akitsugu Kawashima, Tomohiro Yamauchi, Toru Masuoka, Keiichi Akatsuka, Motoi Saito, Makoto Sakamoto, Keiichiro Onitsuka, Masanao Mohri, Oheda Motoki, Akira Machida, Yoshihiko Fu, Masahiko Hiroki, Hitoshi Aiyama, Tomoko Yamana, Kazuki Kobayashi, Norio Ikeda, Nozomu Murai, Hajime Oota, Yuko Tanaka, Satoshi Kuroda, Mikito Hayakawa, Norihiro Sito, Naoyuki Nakao, Hirokatsu Taniguchi, Miyuki Ishikawa, Sonoda Yukihiko, Yuka Itou, Yasuhiro Hamada, Ishihara Hideyuki, Takaaki Inomoto, Yasunobu Mita, Mikito Uchida, Shin Nakano, Hiroshi Ozawa, Takahiro Miyahara, Naoyuki Sakai, Masashi Morikawa, Yasunobu Nakai, Nobuyuki Shimizu, Katsumi Sakata, Masanori Nakagawa, Ichiro Imafuku, Mitsuhiro Iwasaki, Takashi Irioka, Yasuhisa Yoshida, Kinya Nakanishi, and Yoshitaka Suda
References
- 1. United Nations DoEaSA, Population Division . World Urbanization Prospects 2018: Highlights. 2019.
- 2. Buus S, Behrndtz AB, Schmitz ML, Hedegaard JN, Cordsen P, Johnsen SP, Phan T, Andersen G, Simonsen CZ. Urban‐rural inequalities in IV thrombolysis for acute ischemic stroke: a nationwide study. Eur Stroke J. 2024;9:722–731. doi: 10.1177/23969873241244591 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Ramos‐Pachón A, Rodríguez‐Luna D, Martí‐Fàbregas J, Millán M, Bustamante A, Martínez‐Sánchez M, Serena J, Terceño M, Vera‐Cáceres C, Camps‐Renom P, et al. Effect of bypassing the closest stroke Center in Patients with intracerebral hemorrhage: a secondary analysis of the RACECAT randomized clinical trial. JAMA Neurol. 2023;80:1028–1036. doi: 10.1001/jamaneurol.2023.2754 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Garcia‐Tornel A, Millan M, Rubiera M, Bustamante A, Requena M, Dorado L, Olivé‐Gadea M, Jiménez X, Soto A, Querol M, et al. Workflows and outcomes in patients with suspected large vessel occlusion stroke triaged in urban and nonurban areas. Stroke. 2022;53:3728–3740. doi: 10.1161/strokeaha.122.040768 [DOI] [PubMed] [Google Scholar]
- 5. Pérez de la Ossa N, Abilleira S, Jovin TG, García‐Tornel Á, Jimenez X, Urra X, Cardona P, Cocho D, Purroy F, Serena J, et al. Effect of direct transportation to Thrombectomy‐capable center vs local stroke center on neurological outcomes in patients with suspected large‐vessel occlusion stroke in nonurban areas: the RACECAT randomized clinical trial. JAMA. 2022;327:1782–1794. doi: 10.1001/jama.2022.4404 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Loccoh EC, Joynt Maddox KE, Wang Y, Kazi DS, Yeh RW, Wadhera RK. Rural‐urban disparities in outcomes of myocardial infarction, heart failure, and stroke in the United States. J Am Coll Cardiol. 2022;79:267–279. doi: 10.1016/j.jacc.2021.10.045 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Jenks M, Burton E, Williams K. The Compact City: A Sustainable Urban Form?. 1st ed. E & FN Spon; 1996:350. [Google Scholar]
- 8. Ministry of Land I, Transport and tourism, Japan. Towards Form Compact Cities . 2015.
- 9. Kikuchi H, Emberger G, Ishida H, Fukuda A, Kobayakawa S. Dynamic simulations of compact city development to counter future population decline. Cities. 2022;127:103753. doi: 10.1016/j.cities.2022.103753 [DOI] [Google Scholar]
- 10. Glazier RH, Creatore MI, Weyman JT, Fazli G, Matheson FI, Gozdyra P, Moineddin R, Kaufman‐Shriqui V, Booth GL. Density, destinations or both? A comparison of measures of walkability in relation to transportation behaviors, obesity and diabetes in Toronto, Canada. PLoS One. 2014;9:e85295. doi: 10.1371/journal.pone.0085295 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Creatore MI, Glazier RH, Moineddin R, Fazli GS, Johns A, Gozdyra P, Matheson FI, Kaufman‐Shriqui V, Rosella LC, Manuel DG, et al. Association of Neighborhood Walkability with Change in overweight, obesity, and diabetes. JAMA. 2016;315:2211–2220. doi: 10.1001/jama.2016.5898 [DOI] [PubMed] [Google Scholar]
- 12. Howell NA, Tu JV, Moineddin R, Chu A, Booth GL. Association between neighborhood walkability and predicted 10‐year cardiovascular disease risk: the CANHEART (Cardiovascular Health in Ambulatory Care Research Team) cohort. J Am Heart Assoc. 2019;8:e013146. doi: 10.1161/jaha.119.013146 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Koohsari MJ, Nakaya T, Hanibuchi T, Shibata A, Ishii K, Sugiyama T, Owen N, Oka K. Local‐area walkability and socioeconomic disparities of cardiovascular disease mortality in Japan. J Am Heart Assoc. 2020;9:e016152. doi: 10.1161/jaha.119.016152 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Nwana N, Makram OM, Nicolas JC, Pan A, Gullapelli R, Parekh T, Javed Z, Titus A, Al‐Kindi S, Guan J, et al. Neighborhood walkability is associated with lower burden of cardiovascular risk factors among cancer patients. JACC Cardio Oncol. 2024;6:421–435. doi: 10.1016/j.jaccao.2024.03.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Iihara K, Nishimura K, Kada A, Nakagawara J, Ogasawara K, Ono J, Shiokawa Y, Aruga T, Miyachi S, Nagata I, et al. Effects of comprehensive stroke care capabilities on in‐hospital mortality of patients with ischemic and hemorrhagic stroke: J‐ASPECT study. PLoS One. 2014;9:e96819. doi: 10.1371/journal.pone.0096819 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Ren N, Ogata S, Kiyoshige E, Nishimura K, Nishimura A, Matsuo R, Kitazono T, Higashi T, Ogasawara K, Iihara K. Associations between adherence to evidence‐based, stroke quality indicators and outcomes of acute reperfusion therapy. Stroke. 2022;53:3359–3368. doi: 10.1161/strokeaha.121.038483 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Tanimoto R, Hanibuchi T, Nakaya T. Data from: Development of Japan Postcode‐ level Walkability Index and Examination of its Usefulness. 2023. doi: 10.51022/data.tga.22597090.v2. [DOI]
- 18. Gilbert T, Neuburger J, Kraindler J, Keeble E, Smith P, Ariti C, Arora S, Street A, Parker S, Roberts HC, et al. Development and validation of a hospital frailty risk score focusing on older people in acute care settings using electronic hospital records: an observational study. Lancet. 2018;391:1775–1782. doi: 10.1016/s0140-6736(18)30668-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Nakaya T, Honjo K, Hanibuchi T, Ikeda A, Iso H, Inoue M, Sawada N, Tsugane S. Associations of all‐cause mortality with census‐based neighbourhood deprivation and population density in Japan: a multilevel survival analysis. PLoS One. 2014;9:e97802. doi: 10.1371/journal.pone.0097802 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Nieuwenhuijsen MJ. Influence of urban and transport planning and the city environment on cardiovascular disease. Nat Rev Cardiol. 2018;15:432–438. doi: 10.1038/s41569-018-0003-2 [DOI] [PubMed] [Google Scholar]
- 21. Chandrabose M, den Braver NR, Owen N, Sugiyama T, Hadgraft N. Built environments and cardiovascular health: REVIEW AND IMPLICATIONS. J Cardiopulm Rehabil Prev. 2022;42:416–422. doi: 10.1097/hcr.0000000000000752 [DOI] [PubMed] [Google Scholar]
- 22. Bhatnagar A. Environmental determinants of cardiovascular disease. Circ Res. 2017;121:162–180. doi: 10.1161/circresaha.117.306458 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Lai KY, Webster C, Gallacher JE, Sarkar C. Associations of urban built environment with cardiovascular risks and mortality: a systematic review. J Urban Health. 2023;100:745–787. doi: 10.1007/s11524-023-00764-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Ho HC, Guo H, Chan TC, Shi Y, Webster C, Fong KNK. Community planning for a “healthy built environment” via a human‐environment nexus? A multifactorial assessment of environmental characteristics and age‐specific stroke mortality in Hong Kong. Chemosphere. 2022;287:132043. doi: 10.1016/j.chemosphere.2021.132043 [DOI] [PubMed] [Google Scholar]
- 25. Orioli R, Antonucci C, Scortichini M, Cerza F, Marando F, Ancona C, Manes F, Davoli M, Michelozzi P, Forastiere F, et al. Exposure to residential greenness as a predictor of cause‐specific mortality and stroke incidence in the Rome Longitudinal Study. Environ Health Perspect. 2019;127:27002. doi: 10.1289/ehp2854 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Villeneuve PJ, Jerrett M, Su JG, Burnett RT, Chen H, Wheeler AJ, Goldberg MS. A cohort study relating urban green space with mortality in Ontario, Canada. Environ Res. 2012;115:51–58. doi: 10.1016/j.envres.2012.03.003 [DOI] [PubMed] [Google Scholar]
- 27. Monkkonen P, Montejano J, Guerra E, Caudillo C. Compact cities and economic productivity in Mexico. Urban Stud. 2020;57:2080–2097. doi: 10.1177/0042098019869827 [DOI] [Google Scholar]
- 28. Bin Sulaiman F. Compact City: what is the extent of our exploration for its meanings? A systematic review. Sustain For. 2023;15:10302. [Google Scholar]
- 29. Parry‐Jones AR, Sammut‐Powell C, Paroutoglou K, Birleson E, Rowland J, Lee S, Cecchini L, Massyn M, Emsley R, Bray B, et al. An intracerebral hemorrhage care bundle is associated with lower case fatality. Ann Neurol. 2019;86:495–503. doi: 10.1002/ana.25546 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Verburgt E, Verhoeven JI, Hilkens NA, Vaartjes I, De Leeuw F‐E. Age‐specific rural‐urban disparities in the incidence of ischemic stroke in The Netherlands. Neurology. 2024;103:e210102. doi: 10.1212/WNL.0000000000210102 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Kapral MK, Austin PC, Jeyakumar G, Hall R, Chu A, Khan AM, Jin AY, Martin C, Manuel D, Silver FL, et al. Rural‐urban differences in stroke risk factors, incidence, and mortality in people with and without prior stroke. Circ Cardiovasc Qual Outcomes. 2019;12:e004973. doi: 10.1161/circoutcomes.118.004973 [DOI] [PubMed] [Google Scholar]
- 32. Wang W, Jiang B, Sun H, Ru X, Sun D, Wang L, Wang L, Jiang Y, Li Y, Wang Y, et al. Prevalence, incidence, and mortality of stroke in China: results from a Nationwide population‐based survey of 480 687 adults. Circulation. 2017;135:759–771. doi: 10.1161/circulationaha.116.025250 [DOI] [PubMed] [Google Scholar]
- 33. Correia M, Silva MR, Magalhães R, Guimarães L, Silva MC. Transient ischemic attacks in rural and urban northern Portugal: incidence and short‐term prognosis. Stroke. 2006;37:50–55. doi: 10.1161/01.STR.0000195209.26543.8f [DOI] [PubMed] [Google Scholar]
- 34. Jaja BN, Lingsma H, Schweizer TA, Thorpe KE, Steyerberg EW, Macdonald RL. Prognostic value of premorbid hypertension and neurological status in aneurysmal subarachnoid hemorrhage: pooled analyses of individual patient data in the SAHIT repository. J Neurosurg. 2015;122:644–652. doi: 10.3171/2014.10.Jns132694 [DOI] [PubMed] [Google Scholar]
- 35. Thrift AG, McNeil JJ, Forbes A, Donnan GA; Melbourne Risk Factor Study Group . Three important subgroups of hypertensive persons at greater risk of intracerebral hemorrhage. Hypertension. 1998;31:1223–1229. doi: 10.1161/01.hyp.31.6.1223 [DOI] [PubMed] [Google Scholar]
- 36. Cipolla MJ, Liebeskind DS, Chan SL. The importance of comorbidities in ischemic stroke: impact of hypertension on the cerebral circulation. J Cereb Blood Flow Metab. 2018;38:2129–2149. doi: 10.1177/0271678x18800589 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Fujita K, Tanaka K, Yamagami H, Ide T, Ishiyama H, Sonoda K, Satow T, Takahashi JC, Ihara M, Koga M, et al. Detrimental effect of chronic hypertension on leptomeningeal collateral flow in acute ischemic stroke. Stroke. 2019;50:1751–1757. doi: 10.1161/strokeaha.119.025142 [DOI] [PubMed] [Google Scholar]
- 38. Buscot MJ, Chandra RV, Maingard J, Nichols L, Blizzard L, Stirling C, Smith K, Lai L, Asadi H, Froelich J, et al. Association of Onset‐to‐Treatment Time with Discharge Destination, mortality, and complications among patients with aneurysmal subarachnoid hemorrhage. JAMA Netw Open. 2022;5:e2144039. doi: 10.1001/jamanetworkopen.2021.44039 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Davis SM, Broderick J, Hennerici M, Brun NC, Diringer MN, Mayer SA, Begtrup K, Steiner T. Hematoma growth is a determinant of mortality and poor outcome after intracerebral hemorrhage. Neurology. 2006;66:1175–1181. doi: 10.1212/01.wnl.0000208408.98482.99 [DOI] [PubMed] [Google Scholar]
- 40. Morotti A, Boulouis G, Dowlatshahi D, Li Q, Shamy M, Salman RA‐S, Rosand J, Cordonnier C, Goldstein JN, Charidimou A. Intracerebral haemorrhage expansion: definitions, predictors, and prevention. Lancet Neurol. 2023;22:159–171. doi: 10.1016/S1474-4422(22)00338-6 [DOI] [PubMed] [Google Scholar]
- 41. Wang L, Zhou K, Zhang S, Moudon AV, Wang J, Zhu Y‐G, Sun W, Lin J, Tian C, Liu M. Designing bike‐friendly cities: interactive effects of built environment factors on bike‐sharing. Transp Res Part D: Transp Environ. 2023;117:103670. doi: 10.1016/j.trd.2023.103670 [DOI] [Google Scholar]
- 42. Chen Z, Dazard JE, Khalifa Y, Motairek I, Kreatsoulas C, Rajagopalan S, Al‐Kindi S. Deep learning‐based assessment of built environment from satellite images and Cardiometabolic disease prevalence. JAMA Cardiol. 2024;9:556–564. doi: 10.1001/jamacardio.2024.0749 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Liu M, Patel VR, Salas RN, Rice MB, Kazi DS, Zheng Z, Wadhera RK. Neighborhood environmental burden and cardiovascular health in the US. JAMA Cardiol. 2024;9:153–163. doi: 10.1001/jamacardio.2023.4680 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Xie B, Jiao J, An Z, Zheng Y, Li Z. Deciphering the stroke‐built environment nexus in transitional cities: conceptual framework, empirical evidence, and implications for proactive planning intervention. Cities. 2019;94:116–128. doi: 10.1016/j.cities.2019.05.035 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Ren N, Nishimura A, Kurogi A, Nishimura K, Matsuo R, Ogasawara K, Hashimoto Y, Higashi T, Sakai N, Toyoda K, et al. Measuring quality of Care for Ischemic Stroke Treated with Acute Reperfusion Therapy in Japan ‐ the close the gap‐stroke. Circ J. 2021;85:201–209. doi: 10.1253/circj.CJ-20-0639 [DOI] [PubMed] [Google Scholar]
- 46. Nishimura A, Nishimura K, Onozuka D, Matsuo R, Kada A, Kamitani S, Higashi T, Ogasawara K, Shimodozono M, Harada M, et al. Development of quality indicators of stroke centers and feasibility of their measurement using a Nationwide insurance claims database in Japan― J‐ASPECT study ―. Circ J. 2019;83:2292–2302. doi: 10.1253/circj.CJ-19-0089 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
J‐ASPECT Study Collaborators
Data S1
Supplemental Methods
Table S1
Figures S1–S2