Developing Neighborhood Archetypes for Understanding Disparities in Cancer

项目摘要/摘要 癌症存活率的差异仍然存在于种族/民族、社会经济地位(SES)和出生地之间。 邻里因素是形成和延续差异的关键途径。然而， 文献表明，结果喜忧参半，以社区为基础的干预措施收效甚微。至 推进对社区和健康的研究，新的方法需要反映这样一个事实 对健康的影响是通过复杂和动态的过程产生的，强调了学习笔记的重要性 不仅社区特征对健康结果的个体影响，而且还包括它们之间的协同效应。一个 有希望的方法是在单个分类系统中包含多个维度，原型，即 捕捉不同社区的有意义的区别(例如，低SES，农村；贫穷，城市少数民族；中产阶级 郊区/郊区家庭)。原型方法是最大限度地集成多个 关于社区特征的可用信息来源，并有可能确定改进的 提供干预和缩小健康差距的机会。利用一个全面的小型 地区一级的社会和已建成邻里环境数据，并应用新的统计方法，这 应用程序建议在2000年和2010年这两个时间段左右为 加利福尼亚州面积很大、种类繁多的州。中心假设是1)多个邻里因素可以 结合在一起形成邻居原型；以及2)乳腺癌和前列腺癌的死亡率 癌症患者会因这些社区原型而有所不同，这些关联也会因种族/民族而异 和耶稣降生。本研究的具体目的是：1)研究原型教学法的实用性 来描述加州不同的社区。具体地说，我们将扩展到人口普查区 使用确认性潜在类开发的群体级别先前的人口普查区域邻里原型 分析(LCA)。我们还将扩大原有社区，增加建筑和社会环境功能 使用探索性生命周期评价的原型。2)考察邻里原型和邻居原型之间的关联 一组基于人群、种族/民族多样化的乳腺癌和前列腺癌的存活率 病人。具体地说，我们将测试邻里原型是否与所有原因和癌症有关- 具体死亡率以及这种联系是否受到种族/民族的影响。我们还将测试耶稣降生 缓和拉美裔和亚裔美国人之间的邻里原型联系。由此产生的 邻里原型将为未来的研究提供创新的资源。将原型应用于 调查加州不同人群中乳腺癌和前列腺癌的存活率，这项研究将得出 深入了解邻里特征如何相互协同并与个人互动 影响健康的特征。这项研究有可能为更广泛、更多的 基本了解场所对健康的影响，并将其应用于人口健康和健康 差异研究。

PROJECT SUMMARY/ABSTRACT Disparities in cancer survival continue to persist by race/ethnicity, socioeconomic status (SES) and nativity. Neighborhood factors are critical pathways through which disparities are shaped and perpetuated. However, the literature has shown mixed results and neighborhood-based interventions have had limited success. To advance research on neighborhoods and health, new approaches need to reflect the fact that neighborhood effects on health arise through complex and dynamic processes, underscoring the importance of studying not only the individual effects of neighborhood features on health outcomes, but also their synergistic effects. A promising approach is to encompass multiple dimensions within a single classification system, archetypes, that capture meaningful distinctions across neighborhoods (e.g., Low SES, rural; Poor, urban minority; Middle-class suburban/exurban families). The archetype approach is an efficient way to maximally integrate multiple sources of available information on neighborhood characteristics and has the potential to identify improved opportunities for intervening and reducing health disparities. Leveraging a comprehensive database of small- area level data on social and built neighborhood environments and applying novel statistical approaches, this application proposes to develop neighborhood archetypes around two time periods, 2000 and 2010, for the large, diverse state of California. The central hypotheses are that 1) multiple neighborhood factors can be combined together to form neighborhood archetypes; and 2) that cancer mortality among breast and prostate cancer patients will vary by these neighborhood archetypes, and these associations will vary by race/ethnicity and nativity. The specific aims of this study are to: 1) Investigate the utility of the archetypes approach to characterize distinct neighborhoods across California. Specifically, we will extend to the census block group level previously developed census tract neighborhood archetypes using confirmatory latent class analysis (LCA). We will also expand, with additional built and social environment features, prior neighborhood archetypes using exploratory LCA. 2) Examine the associations between neighborhood archetypes and survival among a population-based, racially/ethnically diverse set of breast and prostate cancer patients. Specifically, we will test whether neighborhood archetypes are associated with all cause and cancer- specific mortality and whether this association is moderated by race/ethnicity. We will also test whether nativity moderates the neighborhood archetype associations among Hispanics and Asian Americans. The resulting neighborhood archetypes will provide an innovative resource for future research. Applying the archetypes to investigate breast and prostate cancer survival among California's diverse population, this research will yield insights into how neighborhood characteristics work synergistically with each other and interact with individual characteristics to influence health. This research has the potential to contribute to a broader, more fundamental understanding of place effects on health, with applications to both population health and health disparities research.