Racial Segregation & Disparities in Cancer Outcomes

肺癌、结肠癌、前列腺癌和乳腺癌是美国最常见的癌症。对于每一种常见癌症，有证据表明在诊断阶段、癌症治疗的使用和/或死亡率方面存在持续的种族和民族差异。先前研究癌症差异的工作主要集中在个体特征上，并没有完全解释癌症差异的原因。越来越多的文献表明，一个人居住的地方的属性可能与个人特征无关的健康结果有关。居住隔离是社会环境的特征之一，可能会影响癌症治疗和结果的差异。隔离可能与癌症治疗的使用和结果有关，因为在获得治疗的机会、个人寻求治疗的提供者和设施的特点、对医疗的信念、承受的压力、社会支持、社会经济剥夺以及获得商品和服务的机会等方面存在差异。医疗保健利用行为模型将被用作我们假设的概念框架。来自SEER和Medicare的数据将用于获得诊断为肺癌、结直肠癌、乳腺癌和前列腺癌的个体的人口统计学特征、临床特征和结果。这些个人层面的数据将与有关居住隔离的地区层面数据联系起来。我们将确定种族隔离是否与癌症预后有关，特别是对白人、非洲裔美国人或西班牙裔美国人。关于社会和自然环境特征的其他数据将作为隔离与癌症结果之间关联的潜在中介进行研究。为了检验隔离对癌症结果差异的影响，回归模型将考虑数据的多层次结构。拟议的研究将提供经验证据，可以推动基于地区的干预措施，以减少癌症的差异。

Cancer of the lung, colon, prostate and breast are the most common cancers in the United States. For each of these common cancers, there is evidence of persistent racial and ethnic disparities in stage at diagnosis, the use of cancer treatments, and/or mortality. Prior work examining cancer disparities has focused on individual characteristics, and has not fully explained the causes of cancer disparities. A growing literature suggests that the attributes of the place where an individual resides may be associated with health outcomes independent of individual characteristics. Residential segregation is one of the characteristics of the social environment that is likely to influence disparities in cancer treatments and outcomes. Segregation may be associated with the use of cancer treatments and outcomes through differences in access to care, the characteristics of providers and facilities where an individual seeks care, beliefs about medical treatments, exposure to stress, social support, socioeconomic deprivation, and access to goods and services. The Behavioral Model of Healthcare Utilization will be used as the conceptual framework for our hypotheses. Data from SEER and Medicare will be used to obtain individual demographic characteristics, clinical characteristics, and outcomes for individuals diagnosed with lung, colorectal, breast and prostate cancer. This individual-level data will be linked with area-level data about residential segregation. We will determine whether segregation is associated with cancer outcomes overall, and specifically for individuals who are white, African American or Hispanic. Additional data about the characteristics of the social and physical environment will be examined as potential mediators of the association between segregation and cancer outcomes. To examine the effects of segregation on disparities in cancer outcomes, regression models will account for the multi-level structure of the data. The proposed study would provide empirical evidence that could give impetus to area-based interventions to reduce cancer disparities.