Neighborhoods and cardiovascular risk (RMI)

描述(由申请人提供)：本补充资料“多种族队列中的社区和心血管风险”(R01 HL071759)将建立在父母拨款中开展的工作的基础上，以改进与心血管结果相关的社区水平属性的衡量标准。在快速发展的社区健康影响调查领域，社区水平属性的测量仍然是一个关键挑战。关于不同类型的邻里环境措施的可靠性，以及自我报告的措施与客观措施之间的一致性，缺乏信息。此外，使用多层次技术，根据每个区域的少量观测，为小区域开发改进的邻里级别建筑措施仍处于初级阶段。本附录的目的是(1)调查邻里属性自我报告测量的可靠性(2)。调查基于地理信息系统的测量方法和社区环境自我报告测量方法之间的一致性；(3)调查人口普查数据的使用，以改进使用多水平技术的社区水平属性的经验贝叶斯估计；以及(4)评估替代测量和估计策略对从观察性研究获得的社区健康影响的估计的影响。作为这一方法补充的一部分，将开展的工作将通过(1)量化所用措施的可靠性和(2)改进邻里水平限制的估计，以在将邻里水平因素与台地队列中的健康结果联系起来的模型中用作预测因子，从而与根据母方赠款计划进行的分析直接相关。更广泛地说，拟议的工作将为未来测量邻域水平属性的研究提供参考，方法是提供邻域水平测量的可靠性和有效性的估计，并进一步发展计量经济学技术在测量邻域水平结构方面的应用。

DESCRIPTION (provided by applicant): This supplement to "Neighborhoods and Cardiovascular Risk in a Multiethnic Cohort" (R01 HL071759) will build on work carried out in the parent grant in order to improve the measures of neighborhood-level attributes to be examined in relation to cardiovascular outcomes. The measurement of neighborhood-level attributes remains a key challenge in the rapidly growing field of the investigation of neighborhood health effects. There is scant information on the reliability of different types of measures of neighborhood environments, and on the agreement between self-reported and objective measures. In addition, the use of multilevel techniques to develop improved measures of neighborhood-level constructs for small areas based on small numbers of observations in each area remains in its infancy. The objectives of this supplement are (1) To investigate the reliability of self-reported measures of neighborhood attributes (2). To investigate the agreement between GIS-based measures and self-reported measures of the neighborhood environment (3) To investigate the use of census data to improve empirical-Bayes estimates of neighborhood-level attributes using multilevel techniques; and (4) To evaluate the impact of alternate measurement and estimation strategies on estimates of neighborhood-health effects obtained from observational studies. The work to be carried out as part of this methodological supplement will be of direct relevance to the analyses planned under the parent grant by (1) quantifying the reliability of the measures used and (2) improving the estimates of neighborhood-level constricts to be used as predictors in models relating neighborhood-level factors to health outcomes in the MESA cohort. More generally, the work proposed will inform future studies that measure neighborhood-level attributes by providing estimates of the reliability and validity of neighborhood-level measures and by further developing the application of econometric techniques to the measurement of neighborhood-level constructs.