肺癌
阿卡克信息准则
入射(几何)
医学
人口学
地理变异
普通最小二乘法
环境卫生
流行病学
协变量
回归分析
线性回归
癌症
癌症登记处
回归
疾病
相对风险
空间变异性
人口
空间分析
区域差异
肺病
地理加权回归模型
标准误差
地理
癌症发病率
地理信息系统
作者
Sara M. Myers,Elizabeth Davis,Christopher S. Digesu,Kelly Marie Kenzik
摘要
PURPOSE: A growing share of lung cancers occur in people who have never smoked, and air pollution may be one contributor. Inhalable particulate matter less than 2.5 um in diameter (PM2.5) has been linked to adverse health effects, but its role in lung cancer development across regions of the United States remains unclear. METHODS: Lung cancer incidence from 2018-2022 (via Centers for Disease Control and Prevention) and annual average PM2.5 from 2005-2009 and 2010-2014 (via Area Health Resource File) were studied. Ordinary least squares (OLS) and geographically weighted regression (GWR) models were run, and models were compared using the corrected Akaike information criterion (AICc). Covariates included county-level poverty, racial makeup, educational attainment, and prevalence of smoking. Geographic distribution of PM2.5 levels was mapped to identify spatial clustering of higher and lower exposure (hot and cold spots) using the Getis-Ord Gi* statistic. RESULTS: For both time windows, PM2.5 exposure was positively associated with lung cancer incidence. GWR models demonstrated improved model fit relative to OLS across both PM2.5 exposure periods (2005-2009: AICc 20,892 vs 21,810; 2010-2014: AICc 20,883 vs 21,847). Local GWR coefficient estimates for PM2.5 varied across counties, indicating spatial heterogeneity in the association between PM2.5 exposure and lung cancer incidence. CONCLUSION: The relationship between PM2.5 exposure and lung cancer incidence varies spatially, with different levels of risk in different counties. These exploratory findings show where the association appears stronger or weaker; they are a starting point for further study rather than a basis for county-specific interventions.
科研通智能强力驱动
Strongly Powered by AbleSci AI