Intra-Urban Health Assessed by Remote-Sensing and GIS

描述（由申请人提供）：本项目探索使用遥感图像和地理信息系统，以加纳阿克拉大都市区的数据为研究地点，加强我们对城市内部健康不平等的理解。具体目标如下：(1)通过将空间参考的人口普查数据、调查数据和生命统计数据结合到阿克拉大都市的地理信息系统（GIS）中，得出当地（邻里）的健康措施；(2)通过对遥感影像数据的分类和分析，得出建筑环境和自然环境的局部（邻域）测度；(3)检验城市居民的健康水平受到当地社区环境背景的重要影响，包括自然和建筑环境、社区居民的社会经济构成以及社区在更广泛的城市环境中的位置（包括其与卫生诊所和医院的接近程度）；(4)评估阿克拉城市社区环境背景、人口构成和社区区域属性对健康结果的相对贡献；(5)对预测健康水平的变量之间的相互作用进行建模，以确定可能引入哪些变化以使其整体健康水平达到可接受的最低标准；(6)评估遥感数据本身作为一种代理，在可能导致这些数据被用作健康监测工具的方式下，对城市内部健康不平等进行建模的能力。这项研究包括三个主要步骤：(1)建立数据层并指定要测量的地理参考变量，包括2000年人口普查、1998年和2003年人口与健康调查、2003年阿克拉妇女健康调查、2003年世卫组织加纳世界卫生调查的数据、1999-2001年人口动态统计数据，以及将根据Ridd V-I-S城市生态模型分类的高分辨率多光谱卫星图像；随后，将计算一套景观指标，以衡量建筑环境和自然环境；(2)统计分析，包括空间统计、回归和多层次方法，以回答我们的具体目标提出的问题；(3)结果的解释和传播。

DESCRIPTION (provided by applicant): This project explores the use of remotely sensed imagery and GIS to enhance our understanding of intraurban inequalities in health, using data for the metropolitan area of Accra, Ghana as a study site. The specific aims are as follows: (1) to derive local (neighborhood) measures of health by combining spatially referenced census data, survey data, and vital statistics into a geographic information system (GIS) for metropolitan Accra; (2) to derive local (neighborhood) measures of the built and natural environment through the classification and analysis of data from remotely sensed imagery; (3) to test the hypothesis that health levels in urban places are importantly influenced by the local neighborhood environmental context, including the natural and built environment, the socio-economic composition of the neighborhood's residents, and the location of a neighborhood within the broader urban environment (including its proximity to health clinics and hospitals); (4) to assess the relative contribution of neighborhood environmental context, population composition, and the neighborhood Iocational attributes to health outcomes in metropolitan Accra; (5) to model the interaction among the variables that predict health levels to determine what changes might be introduced into a neighborhood to bring its overall level of health up to a minimally acceptable standard; and (6) to evaluate how well the remotely-sensed data can, on their own as a proxy, model the intra-urban inequalities in health in ways that might lead these data to be used as health monitoring tools. The research involves three major steps: (1) creation of data layers and specification of georeferenced variables to be measured, including data from the 2000 Census, the 1998 and 2003 Demographic and Health Surveys, the 2003 Women's Health in Accra Survey, the 2003 WHO World Health Survey in Ghana, data from the vital statistics for 1999-2001, and a high resolution multispectral satellite image which will be classified according to the Ridd V-I-S model of urban ecology, subsequent to which a set of landscape metrics will be calculated to measure the built and natural environments; (2) statistical analysis to answer the questions posed by our Specific Aims, which will include spatial statistics, regression, and multi-level approaches; and (3) interpretation and dissemination of the results.