Objectives: To explore the spatial distribution characteristics, trend changes, and spatial clustering of esophageal cancer among residents in China at the county (city, district) scale, a spatial epidemiological approach was used, with the aim of providing localized evidence for the prevention and treatment of esophageal cancer in China. Methods: The data source was the incidence (crude rate) and mortality (crude rate) of esophageal cancer from 2005 to 2016 in the 2008-2019 edition of China Cancer Registration Annual Report published by the National Cancer Center. The Joinpoint model was used for time trend analysis. The tumor registration area in 2016 was selected as the study area for spatial feature analysis, with a total of 487 counties (cities and districts), covering 27.6% of the national population. Spatial autocorrelation analysis was performed to reveal spatial distribution characteristics by using Arcgis 10.6 software, and spatial scanning statistics was used to analyze spatial clustering characteristics by using SaTScan 9.5 software. The log-likelihood ratio (LLR) and relative risk (RR) were calculated in different windows, and the region with the largest LLR value represented the most likely cluster. Results: From 2005 to 2016, the incidence and mortality rate of esophageal cancer in China showed a trend of increasing at first and then decreasing. The incidence and mortality rate of esophageal cancer in 2016 were characterized by spatial positive correlation. High incidence and high mortality were mainly concentrated in the areas through which the Huaihe River flowed. The primary clusters (taking high incidence rate as an example LLR=6 374.41, RR=2.37, P＜0.001) were mainly distributed in Jiangsu, Anhui and Shandong in eastern China and eastern Henan and southern Hebei in central China, and secondary clusters (taking high incidence rate as an example LLR=1 971.19, RR=1.91, P＜0.001) in Gansu, Ningxia Hui Autonomous Region, Shaanxi, Sichuan and other central and western regions. Conclusions: The incidence and mortality of esophageal cancer in China have decreased since 2010. The disease burden of esophageal cancer has obvious spatial differences, and measures should be taken according to local conditions in high-risk cluster areas such as the Huaihe River basin.
目的： 采用空间流行病学方法，以县（市、区）为尺度，探索中国居民食管癌发病率和死亡率的时间变化趋势和空间分布特征，为食管癌的防治工作提供本土化证据。 方法： 数据来源为国家癌症中心发布的2008—2019版《中国肿瘤登记年报》中2005—2016年食管癌发病率（粗率）和死亡率（粗率），采用Joinpoint模型进行时间趋势分析。选取2016年肿瘤登记地区作为空间特征分析的研究区域，共计487个县（市、区），覆盖27.6%的中国人口数。使用Arcgis 10.6软件进行空间自相关分析揭示空间分布特征，使用SaTScan 9.5软件进行空间扫描统计分析空间聚类特征，计算不同窗口下的对数似然比（LLR）和相对危险度（RR），LLR值最大的区域代表最有可能的聚类。 结果： 2005—2016年中国食管癌发病率和死亡率呈现先升后降的趋势。2016年食管癌的发病率和死亡率具有空间正相关性特征，高发病率和高死亡率主要聚集在淮河流经的地区，一级聚集区（高发病率聚集区LLR=6 374.41，RR=2.37，P＜0.001）主要分布在华东地区的江苏省、安徽省、山东省和华中地区的河南省东部、河北省南部，二级聚集区（高发病率聚集区LLR=1 971.19，RR=1.91，P＜0.001）主要分布在甘肃省、宁夏回族自治区、陕西省、四川省等中西部地区。 结论： 2010年以来中国食管癌发病率和死亡率有所下降。食管癌疾病负担具有明显的空间差异性，对于淮河流域等高危聚集区应因地制宜采取措施。.