Department of Biostatistics, Virginia Commonwealth University, Richmond, VA 23298, USA.
Department of Health Behavior and Policy, Virginia Commonwealth University, Richmond, VA 23298, USA.
Int J Environ Res Public Health. 2022 Apr 26;19(9):5244. doi: 10.3390/ijerph19095244.
More than 30% of cancer related deaths are related to tobacco or alcohol use. Controlling and restricting access to these cancer-causing products, especially in communities where there is a high prevalence of other cancer risk factors, has the potential to improve population health and reduce the risk of specific cancers associated with these substances in more vulnerable population subgroups. One policy-driven method of reducing access to these cancer-causing substances is to regulate where these products are sold through the placement and density of businesses selling tobacco and alcohol. Previous work has found significant positive associations between tobacco, alcohol, and tobacco and alcohol retail outlets (TRO, ARO, TARO) and a neighborhood disadvantage index (NDI) using Bayesian shared component index modeling, where NDI associations differed across outlet types and relative risks varied by population density (e.g., rural, suburban, urban). In this paper, we used a novel Bayesian index model with spatially varying effects to explore spatial nonstationarity in NDI effects for TROs, AROs, and TAROs across census tracts in North Carolina. The results revealed substantial variation in NDI effects that varied by outlet type. However, all outlet types had strong positive effects in one coastal area. The most important variables in the NDI were percent renters, Black racial segregation, and the percentage of homes built before 1940. Overall, more disadvantaged areas experienced a greater neighborhood burden of outlets selling one or both of alcohol and tobacco.
超过 30%的与癌症相关的死亡与烟草或酒精使用有关。控制和限制这些致癌产品的获取，特别是在其他癌症风险因素高发的社区，有可能改善人口健康，并降低与这些物质相关的特定癌症在更脆弱人群亚组中的风险。一种通过政策来减少这些致癌物质获取的方法是通过对销售烟草和酒精的企业的位置和密度来规范这些产品的销售地点。以前的工作使用贝叶斯共享分量指数模型发现了烟草、酒精和烟草及酒精零售点（TRO、ARO、TARO）与邻里劣势指数（NDI）之间存在显著的正相关关系，其中 NDI 的关联因销售点类型而异，相对风险因人口密度而异（例如，农村、郊区、城市）。在本文中，我们使用了一种具有空间变化效应的新颖的贝叶斯指数模型，来探索北卡罗来纳州普查区内 TRO、ARO 和 TARO 的 NDI 效应的空间非平稳性。结果显示，NDI 效应存在很大差异，且因销售点类型而异。然而，所有销售点类型在一个沿海地区都有强烈的正效应。NDI 中最重要的变量是租房者的比例、黑人种族隔离程度和 1940 年前建造的房屋比例。总体而言，处于劣势地位的地区更容易受到销售一种或两种酒精和烟草的销售点的邻里负担的影响。
