Exploring neighbourhood effects, genetic characteristics and individual behaviours' influence on health disparities within a geographical context...

…基于社交媒体数据挖掘和个人层面的调查数据。肥胖症和糖尿病等慢性疾病的流行率迅速上升，已成为世界范围内的一项重大挑战和威胁，需要更多的关注和潜在的干预措施。人类的健康状况不可避免地受到生活环境的影响，包括建筑、社会、经济和政治环境、个人特征和行为以及遗传变异。本研究主要关注健康差异的地理变异，以及地理环境如何在环境、基因组和个体行为之间的相互作用中发挥重要作用，全面了解健康不平等的潜在机制，调查慢性疾病的危险因素。利用个体水平的调查数据，基于不同的探索性统计和回归模型，探索、识别和分类邻里效应和遗传对健康状况的影响。此外，互联网的广泛使用提供了分享日常生活的机会，也提供了观察健康相关行为的机会，如健康食品消费、体育活动、吸烟、饮酒和不良睡眠模式（Pachucki等人，2011；Keating等人，2011；Rosenquist等人，2010；Mednick等人，2010）。通过社交媒体数据挖掘，可以在更广泛的人群中观察到更多的健康行为模式，从而为调查社会文化因素与健康结果之间的关系提供了机会；还将根据调查数据和与健康有关的行为（如从社交媒体平台获取的饮食习惯和体育活动）相结合，确定跨空间慢性病的邻里效应和风险因素。分析结果将有助于深入了解邻里效应、遗传特征和个人行为如何影响健康状况，鼓励各国政府提高对不同类型风险因素的认识，并改善旨在解决慢性病高风险问题的生活环境。

... based on social media data mining and individual level survey data. Rapidly increasing prevalence of chronic diseases, such as obesity and diabetes has become a major worldwide challenge and threat requiring more attention and potential interventions. The health status of humans is inevitably affected by the living environment, including the built, social, economic and political environments, their individual characteristics and behaviours, and genetic variations. This research mainly focuses on geographical variations of health disparities and how geographical context plays the significant role in interactions between environments, genomic and individual behaviours, comprehensively understanding the underlying mechanisms of health inequality and investigating risk factors of chronic diseases. Individual level survey data will be utilized to explore, identify and classify the neighbourhood effect and genetic influence on health status based on different exploratory statistical and regression models. Additionally, widespread usage of the Internet provides an opportunity to share daily life, as well as an opportunity to observe health-related behaviours, such as healthy food consumption, physical activity, smoking, alcohol consumption, and poor sleep patterns (Pachucki, et al., 2011; Keating, et al., 2011; Rosenquist, et al., 2010; Mednick, et al., 2010). More health-behaviour patterns can be observed through social media data mining among the wider population, providing an opportunity to investigate the association between socio-cultural factors and health outcomes; also to identity the neighbourhood effect and risk factors of chronic diseases across space based on the combination of survey data and health-related behaviours, such as diet habits and physical activities derived from social media platforms. The analytical results will provide insight on how neighbourhood effect, genetic characteristics and individual behaviour have effects on health status, encouraging governments to raise the awareness of different types of risk factors and improve the living environment aiming at addressing high risks of chronic diseases.