Penney T L, Rainham D G C, Dummer T J B, Kirk S F L
Applied Research Collaborations for Health (ARCH), School of Health and Human Performance, Faculty of Health Professions, Dalhousie University, Halifax, NS, Canada.
J Hum Nutr Diet. 2014 Apr;27 Suppl 2:65-74. doi: 10.1111/jhn.12055. Epub 2013 May 24.
Rates of overweight and obesity are now considered to be epidemic. Few studies have examined the spatial distribution of overweight and obesity at the community level, an area of geography recommended for prevention and intervention. Therefore, the present study aimed to examine the spatial variation of overweight and obesity using community geographic boundaries.
A cross-sectional secondary spatial data analysis was conducted using three combined cycles of Canadian Community Health Survey data for the province of Nova Scotia with community level boundaries. Descriptive rates were calculated using standardised incidence ratio values and spatial analysis was carried out using Global and Local Moran's I and the GetisOrdGi* statistic for cluster identification.
Maps illustrating local cluster analysis showed a significant degree of similarity between neighbouring communities in urban areas more so than rural communities. Hot spot analysis maps showed communities clustering together in the urban centre tended to have lower incidence of overweight and obesity ('cool spots'), whereas clustered communities in a more rural area had a higher incidence of overweight and obesity ('hot spots).
The present study showed that there was geographical variation in overweight and obesity between urban and rural communities, and also there was a tendency for communities to cluster based on the incidence of overweight and obesity. This highlights the importance of understanding community level obesity rates and associated behavioural determinants, such as diet and physical activity, as well as the role that urbanisation or rurality may play in intervention initiatives for these behavioural determinants. Specifically, public health nutrition efforts for community level food environments in rural areas should ensure an individualised approach is used, whereas urban areas may be amenable to more general approaches aiming to support healthy weight status among the broader population.
超重和肥胖率如今被视为一种流行病。很少有研究在社区层面考察超重和肥胖的空间分布情况，而这一地理领域被推荐用于预防和干预。因此，本研究旨在利用社区地理边界来考察超重和肥胖的空间差异。
使用新斯科舍省三个合并周期的加拿大社区健康调查数据进行横断面二次空间数据分析，数据具有社区层面的边界。使用标准化发病率比值计算描述性比率，并使用全局和局部莫兰指数以及GetisOrdGi*统计量进行空间分析以识别聚类。
展示局部聚类分析的地图显示，城市地区相邻社区之间的相似程度明显高于农村社区。热点分析地图显示，聚集在城市中心的社区往往超重和肥胖发病率较低（“冷点”），而农村地区聚集的社区超重和肥胖发病率较高（“热点”）。
本研究表明，城乡社区之间超重和肥胖存在地理差异，而且社区也有基于超重和肥胖发病率进行聚类的趋势。这凸显了了解社区层面肥胖率及相关行为决定因素（如饮食和身体活动）的重要性，以及城市化或农村化在这些行为决定因素干预举措中可能发挥的作用。具体而言，针对农村地区社区层面食物环境的公共卫生营养工作应确保采用个性化方法，而城市地区可能更适合采用旨在支持更广泛人群健康体重状况的更通用方法。
