Health, Poverty and Place: Modeling Inequalities in Accra Using RS and GIS

描述（由申请人提供）：本项目通过对加纳阿克拉城市健康空间差异的详细分析，旨在为解释贫穷国家城市的城市健康不平等提供一个模型。以加纳的阿克拉为例，我们将产生一些方法，我们希望这些方法将成为对发展中国家其他主要城市进行比较分析的模板。该项目将编制城市地区保健和死亡率差别的地理概况，其格式可供政府和负责全面改善保健和减少保健不平等现象的发展机构立即使用；绘制城市贫困（收入和财富）分布地图，并评估其与选定的健康和死亡率结果的空间关联；利用遥感和地理信息系统技术，衡量有害健康和死亡率结果与邻里生态的关系，包括与从多光谱高空间分辨率卫星图像中可见的物理特征和建筑结构观测得出的空间措施的关系；并评估包括社会组织和机构在内的社区一级变量对产生健康和死亡率不平等的额外贡献，不考虑个人一级的特征。我们的研究包括以下主要步骤：(i)应用新的方法，利用人口普查和调查数据以及遥感图像信息，跨空间和时间定义城市街区；(2)在阿克拉有目的的社区样本中，通过实地评估验证数据派生的社区定义、属性和边界；(3)基于衍生的社区定义，以及现有和更新的图像和地理层，在社区一级收集现有的基于人口普查、调查和人口动态统计的发病率和死亡率指标；(4)对2003年阿克拉妇女健康调查的受访者进行重新访谈（2008年），以产生关于社区一级健康结果及其变化的新措施；(5)运用空间分析、多变量和多层次分析预测阿克拉社区之间和社区内部健康结果的不平等；(6)开发基于WHS和DHS数据并结合图像的模型，以预测社区的健康结果；(7)将建模算法应用于两个关键资源-人口与健康调查和遥感图像-来预测阿克拉社区一级健康结果的变异性（不平等）；(8)将结果用于阿克拉当地的卫生政策和规划目的，并利用数据相对丰富的阿克拉的结果开发模型，随后可用于估计撒哈拉以南非洲数据匮乏城市的卫生空间变异性。

DESCRIPTION (provided by applicant): This project, through a detailed analysis of spatial disparities in health in Accra, Ghana, aims to provide a model for the interpretation of urban health inequalities in cities in poor countries. Using Accra in Ghana, we will generate methods that we hope will serve as templates for comparative analyses of other major cities in the developing world. The project will produce a geographical synopsis of the differentials in health and mortality within the urban area in a format immediately accessible to government and the development agencies tasked with the improvement of health overall and with the reduction of health inequalities; map the distribution of poverty (both income and wealth) in the city and assess its spatial association with selected health and mortality outcomes; using remote sensing and GIS technology, measure the association of adverse health and mortality outcomes with neighborhood ecology, including the association with spatial measures derived from observations of physical features and built structures visible from multi-spectral high spatial resolution satellite imagery; and assess the additional contribution of community-level variables including social organization and institutions to the production of inequalities in health and mortality, net of individual-level characteristics. Our research involves the following major steps: (i) apply new methods of defining urban neighborhoods across space and time using census and survey data and information from remotely sensed imagery; (2) validate data-derived neighborhood definitions, attributes, and boundaries with field assessments in a purposive sample of Accra neighborhoods; (3) assemble existing census-based, survey-based, and vital statistics-based measures of morbidity and mortality at the neighborhood level, based on the derived neighborhood definitions, as well as existing and updated imagery and geographic layers; (4) re-interview (in 2008) the respondents to the 2003 Accra Women's Health Survey to generate new measures of neighborhood-level health outcomes and changes in those outcomes; (5) employ spatial analytic, multivariate and multi-level analyses to predict the inequalities in health outcomes between and within neighborhoods of Accra; (6) develop models that build on the WHS and DHS data in combination with the imagery to predict health outcomes by neighborhood; (7) apply the modeling algorithms to two key resources-the Demographic and Health Survey and remotely sensed imagery-to predict the variability (inequality) in neighborhood level health outcomes in Accra; and (8) employ the results locally in Accra for health policy and planning purposes, and use the results from relatively data-rich Accra to develop models that can subsequently be used to estimate the spatial variability in health in data-poorer cities in sub-Saharan Africa.