Developing Neighborhood Archetypes for Understanding Disparities in Cancer

描述（由申请人提供）：癌症生存的差异继续因种族/民族、社会经济地位（SES）和出生而存在。邻里因素是差异形成和延续的关键途径。然而，文献显示的结果好坏参半，以社区为基础的干预措施取得了有限的成功。为了推进社区与健康的研究，新的方法需要反映社区对健康的影响是一个复杂和动态的过程，强调不仅要研究社区特征对健康结果的个体影响，还要研究它们的协同效应。一种很有前景的方法是在一个单一的分类系统中包含多个维度，即原型，它可以捕捉到社区之间有意义的区别（例如，低社会经济地位，农村；贫穷，城市少数民族；中产阶级郊区/远郊家庭）。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。利用关于社会和建筑社区环境的小区域级数据的综合数据库，并应用新颖的统计方法，该应用程序建议在2000年和2010年两个时期为加利福尼亚州的大而多样的州开发社区原型。中心假设是：1)多种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会因这些社区原型而异，这些关联会因种族/民族和出生地而异。本研究的具体目的是：1)调查原型方法的效用，以表征加州不同的社区。具体而言，我们将使用验证性潜在类分析（LCA）扩展到先前开发的普查区社区原型的人口普查区群体水平。我们还将使用探索性LCA扩展额外的建筑和社会环境特征，先前的社区原型。2)在以人群为基础、种族/民族多样化的乳腺癌和前列腺癌患者中，研究社区原型与生存率之间的关系。具体而言，我们将测试邻里原型是否与所有原因和癌症特异性死亡率相关，以及这种关联是否受到种族/民族的调节。我们还将测试出生是否会调节西班牙裔和亚裔美国人的邻里原型关联。由此产生的社区原型将为未来的研究提供创新资源。应用原型来调查乳腺癌和前列腺癌在加州不同人群中的生存情况，这项研究将深入了解社区特征如何相互协同作用，并与个体特征相互作用，从而影响健康。这项研究有可能有助于更广泛、更基本地了解地点对健康的影响，并将其应用于人口健康和健康差异研究。

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.