Developing Neighborhood Archetypes for Understanding Disparities in Cancer

描述(由申请人提供)：癌症存活率的差异继续存在于种族/民族、社会经济地位(SES)和出生地。邻里因素是形成和延续差异的关键途径。然而，文献显示结果喜忧参半，以社区为基础的干预措施收效甚微。为了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，即社区对健康的影响是通过复杂和动态的过程产生的，强调不仅要研究社区特征对健康结果的个别影响，而且还要研究它们的协同效应。一个有希望的方法是在一个单一的分类系统中涵盖多个维度，原型，捕捉不同社区的有意义的区别(例如，社会经济地位低的农村家庭；贫穷的城市少数民族；中产阶级郊区/郊区家庭)。原型方法是最大限度地整合关于社区特征的多种可用信息来源的有效方法，并有可能确定干预和减少健康差距的更好机会。该应用程序利用关于社会和已建社区环境的小区域级别数据的综合数据库，并应用新的统计方法，建议在2000和2010两个时间段左右开发社区原型，适用于面积庞大、多样化的加利福尼亚州。中心假设是：1)多个邻里因素可以组合在一起形成邻里原型；2)乳腺癌和前列腺癌患者的癌症死亡率将因这些邻里原型而异，这些关联将因种族/民族和原籍而异。这项研究的具体目的是：1)调查原型方法在描述加州不同社区特征方面的效用。具体地说，我们将使用验证性潜在类分析(LCA)将先前开发的人口普查区域邻里原型扩展到人口普查区块组级别。我们还将使用探索性LCA扩展先前的社区原型，并增加建筑和社会环境功能。2)在一组以人群为基础、种族/民族多样化的乳腺癌和前列腺癌患者中，研究邻里原型与生存之间的关联。具体地说，我们将测试邻里原型是否与所有原因和癌症特有的死亡率相关，以及这种关联是否受到种族/民族的影响。我们还将测试出生地是否会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。将这些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，这项研究将深入了解社区特征如何相互协同作用，并与个人特征相互作用，以影响健康。这项研究有可能有助于更广泛、更基本地了解地区对健康的影响，并将其应用于人口健康和健康差距研究。

DESCRIPTION (provided by applicant): 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 utiity 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.