Department of Medical Statistics and Epidemiology, School of Public Health, Sun Yat-sen University, People's Republic of China.
Department of Infectious Diseases, Foshan Center for Disease Control and Prevention, People's Republic of China.
Sci Total Environ. 2020 Jul 1;724:138200. doi: 10.1016/j.scitotenv.2020.138200. Epub 2020 Mar 24.
Dengue fever (DF) is a common and rapidly spreading vector-borne viral disease in tropical and subtropical regions. In recent years, in China, DF still poses an increasing threat to public health in many cities; but the incidence shows significant spatiotemporal differences. The purpose of this study was to identify the key factors affecting the spatial and temporal distribution of DF. We collected natural environmental and socio-economic data for two adjacent cities, Guangzhou (73 variables) and Foshan (71 variables), with the most DF cases in China. We performed random forest modelling to rank the factors according to their level of importance, and used negative binomial regression analysis to compare the risk factors between outbreak years and non-outbreak years. The natural environmental factors contributing to DF incidence for Guangzhou were temperature (relative risk (RR) = 18.80, 95% confidence interval (CI) = 3.11-113.67), humidity (RR = 1.85, 95% CI = 1.17-2.90) and green area (RR = 12.11, 95% CI = 6.14-55.50), and for Foshan was forest coverage (RR = 5.83, 95% CI = 2.72-12.45). Socio-economic impact were shown in Guangzhou with foreign visitor (RR = 1.18, 95% CI = 1.05-1.34) and oversea air passenger transport (RR = 7.34, 95% CI = 2.26-23.86); in Foshan, with oversea tourism (RR = 1.65, 95% CI = 1.34-2.04); and in Guangzhou-Foshan, with the development of intercity metro (RR = 1.26, 95% CI = 1.10-1.44). The difference between the two cities was the greater impact of the foreign visitor, green spaces and climate factor on DF in Guangzhou; the impact of the construction of intercity metro; and the more significant impact of oversea tourism on DF in Foshan. Our results suggest meaningful clues to public health authorities implementing joint interventions on DF prevention and early warning, to increase health education on DF prevention for international visitors and oversea travelers, and cross-city metro passengers; using rapid body temperature detection in public places such as airports, metros and parks can help detect cases early.
登革热（DF）是一种在热带和亚热带地区常见且迅速传播的虫媒病毒性疾病。近年来，在中国，DF 仍对许多城市的公共卫生构成日益严重的威胁；但发病率呈现出显著的时空差异。本研究旨在确定影响 DF 时空分布的关键因素。我们收集了中国两个相邻城市——广州（73 个变量）和佛山（71 个变量）的自然环境和社会经济数据，这些城市的 DF 病例最多。我们使用随机森林模型根据重要性对因素进行排名，并使用负二项回归分析比较了爆发年份和非爆发年份的风险因素。对广州 DF 发病率有影响的自然环境因素包括温度（相对风险 (RR) = 18.80，95%置信区间 (CI) = 3.11-113.67）、湿度（RR = 1.85，95% CI = 1.17-2.90）和绿地面积（RR = 12.11，95% CI = 6.14-55.50），对佛山的影响因素是森林覆盖率（RR = 5.83，95% CI = 2.72-12.45）。在广州，社会经济影响表现为外国游客（RR = 1.18，95% CI = 1.05-1.34）和海外航空旅客运输（RR = 7.34，95% CI = 2.26-23.86）；在佛山，海外旅游（RR = 1.65，95% CI = 1.34-2.04）；在广佛地区，城市间地铁的发展（RR = 1.26，95% CI = 1.10-1.44）。两个城市的区别在于，外国游客、绿地和气候因素对广州 DF 的影响更大；对城际地铁建设的影响；以及海外旅游对佛山 DF 的影响更为显著。我们的研究结果为公共卫生部门实施 DF 预防和预警的联合干预措施提供了有意义的线索，为国际游客和海外旅行者以及跨城市地铁乘客增加了 DF 预防健康教育；在机场、地铁和公园等公共场所使用快速体温检测有助于早期发现病例。
