Supplementary Material for: An Ecological Study of the Association between Air Pollution and Hepatocellular Carcinoma Incidence in Texas

<strong><em>Introduction:</em></strong> Primary liver cancer is a significant cause of cancer-related death in both the United States and the world at large. Hepatocellular carcinoma comprises 90% of these primary liver cancers and has numerous known etiologies. Evaluation of these identified etiologies and other traditional risk factors cannot explain the high incidence rates of hepatocellular carcinoma in Texas. Texas is home to the second largest petrochemical industry and agricultural industry in the nation; industrial activity and exposure to pathogenic chemicals have never been assessed as potential links to the state's increased incidence rate of hepatocellular carcinoma. <b><i>Methods:</i></b> The association between the county-level concentrations of 4 air pollutants known to be linked to liver cancer, vinyl chloride, arsenic, benzene, and 1,3-butadiene, and hepatocellular carcinoma rates was evaluated using nonparametric generalized additive logistic regression and gamma regression models. Hepatocellular carcinoma incidence rates for 2000-2013 were evaluated in comparison to 1996 and 1999 pollution concentrations and hepatocellular carcinoma rates for the subset of 2006-2013 were evaluated in comparison to 2002 and 2005 pollution concentrations, respectively. <b><i>Results:</i></b> The analysis indicates that the relationship between the incidence of liver cancer and air pollution and risk factors is nonlinear. There is a consistent significant positive association between the incidence of liver cancer and hepatitis C prevalence rates (gamma all years, <i>p</i> &lt; 0.05) and vinyl chloride concentrations (logistic 2002 and 2005, <i>p</i> &lt; 0.0001; gamma 2002 and 2005, <i>p</i> &lt; 0.05). <b><i>Conclusions:</i></b> This study suggests that vinyl chloride is a significant contributor to the incidence of liver cancer in Texas. The relationship is notably nonlinear. Further, the study supports the association between incidence of liver cancer and prevalence of hepatitis B.

<strong><em>引言：</em></strong> 原发性肝癌（Primary liver cancer）是美国乃至全球范围内与癌症相关死亡的重要诱因。肝细胞癌（Hepatocellular carcinoma）占此类原发性肝癌的90%，已知存在多种致病病因。针对上述已明确的病因及其他传统风险因素的评估，无法解释德克萨斯州肝细胞癌的高发病率。德克萨斯州拥有全美第二大石油化工产业与农业产业，而工业活动及致病化学物质暴露情况，此前从未被作为该州肝细胞癌发病率上升的潜在关联因素开展研究。
<strong><em>方法：</em></strong> 本研究针对4种已知与肝癌相关的空气污染物——氯乙烯（vinyl chloride）、砷（arsenic）、苯（benzene）及1,3-丁二烯（1,3-butadiene）的县级浓度与肝细胞癌发病率之间的关联，采用非参数广义加性logistic回归与gamma回归模型进行评估。针对2000-2013年的肝细胞癌发病率，以1996年与1999年的污染物浓度作为对照进行分析；针对2006-2013年的子数据集肝细胞癌发病率，则分别以2002年与2005年的污染物浓度作为对照开展评估。
<strong><em>结果：</em></strong> 分析表明，肝癌发病率与空气污染及风险因素之间的关联呈非线性特征。肝癌发病率与丙型肝炎（hepatitis C）患病率（全年度gamma模型分析，<i>p</i> < 0.05）以及氯乙烯浓度之间存在一致且显著的正相关关系：logistic回归分析（2002年与2005年数据集）结果显示<i>p</i> < 0.0001；gamma回归分析（2002年与2005年数据集）结果显示<i>p</i> < 0.05。
<strong><em>结论：</em></strong> 本研究提示氯乙烯是德克萨斯州肝癌发病率上升的重要贡献因素，二者的关联呈显著非线性特征。此外，本研究支持肝癌发病率与乙型肝炎（hepatitis B）患病率之间存在关联。