A pan-Canadian measure of active living environments using open data
A pan-Canadian measure of active living environments using open data
复制标题
DOI：
10.25318/82-003-x201900500002-eng
复制
发表时间：
2019-05-01
期刊：
影响因子：
5
通讯作者：
Ross, Nancy A.
中科院分区：
文献类型：
--
作者：
Herrmann, Thomas;Gleckner, William;Ross, Nancy A.
背景：支持积极生活（如步行或骑自行车出行）的社区环境可能会减轻与久坐行为相关的慢性病负担。有许多方法可用于总结支持积极生活的社区特征，但很少有适用于全加拿大的，且没有一种使用可广泛共享的开放数据源。本研究报告了一套利用开放数据开发和验证的新型积极生活环境指标，这些数据可与全国健康调查相关联，并可供地方、地区或国家政府用于公共卫生监测。
数据与方法：使用地理信息系统（GIS）计算了加拿大各地56589个传播区域（DA，2016年数据）的连通性、密度以及与目的地的接近程度等多种指标。计算皮尔逊相关系数以评估每项指标与人口普查数据中步行上班和采用积极交通方式上班（步行、骑自行车和使用公共交通的组合）的比率之间的关联。为最终数据库选择的积极生活环境指标通过k - 中位数聚类法，根据积极生活环境的有利程度将传播区域分类。
结果：在传播区域层面，所有指标都与步行上班和积极交通方式上班的比率相关，无论它们是使用专有数据源还是开放数据源得出的。开放数据的覆盖范围在加拿大各地区是一致的。基于相关性分析以及适用于各种社区规模和公开可用数据的原则，为加拿大积极生活环境（Can - ALE）数据集选择了三项指标：（1）源自开放街道地图（OSM）的道路和人行道三岔路口密度，（2）源自加拿大统计局住宅数量的加权住宅密度，（3）源自OSM的兴趣点。对于较大城市地区的传播区域子集，增加了一项公共交通可达性指标，且该指标与积极交通方式上班的比率密切相关。通过聚类分析得出的五个Can - ALE组逐步对积极交通方式上班的比率进行了分级，尽管只有在最有利的积极生活环境中，步行上班的比率才超过全国平均水平。
解释：开放数据可用于得出描述加拿大社区积极生活环境特征的指标。
Background: Neighbourhood environments that support active living, such as walking or cycling for transportation, may decrease the burden of chronic conditions related to sedentary behaviour. Many measures exist to summarize features of communities that support active living, but few are pan-Canadian and none use open data sources that can be widely shared. This study reports the development and validation of a novel set of indicators of active living environments using open data that can be linked to national health surveys and can be used by local, regional or national governments for public health surveillance.Data and methods: A Geographic Information System (GIS) was used to calculate a variety of measures of the connectivity, density and proximity to destinations for 56,589 dissemination areas (DAs) across Canada (2016 data). Pearson correlation coefficients were calculated to assess the association between each measure and the rates of walking to work and taking active transportation to work (a combination of walking, cycling and using public transportation) from census data. The active living environment measures selected for the final database were used to classify the DAs by the favourability of their active living environment into groups by k-medians clustering.Results: All measures were correlated with walking-to-work and active-transportation-to-work rates at the DA level, whether they were derived using proprietary or open data sources. Coverage of open data was consistent across Canadian regions. Three measures were selected for the Canadian Active Living Environments (Can-ALE) dataset based on the correlation analysis, but also on the principles of suitability for a variety of community sizes and openly available data: (1) three-way intersection density of roads and footpaths derived from OpenStreetMap (OSM), (2) weighted dwelling density derived from Statistics Canada dwelling counts and (3) points of interest derived from OSM. A measure of access to public transportation was added for the subset of DAs in larger urban areas and was strongly related to active-transportation-to-work rates. Active-transportation-to-work rates were graded, in steps, by the five Can-ALE groups derived from the cluster analysis, although walking-to-work rates exceeded the national average only in the most favourable active living environments.Interpretation: Open data may be used to derive measures that characterize the active living environments of Canadian communities.