Abstract: Background　Dengue is the fastest-rising mosquito-borne disease in China, with its prevalence rapidly expanding northward in recent years. Objective　To analyze the disease burden, spatiotemporal patterns, and determinants of Dengue in China from 2004 to 2023. Methods　Data from the Global Burden of Disease Study 2024, Chinese Public Health Science Data Center, China Health Statistics Yearbook, National Statistics, and weather databases were utilized. Estimated annual percentage change (EAPC) for incidence and mortality were calculated. Spatiotemporal distribution and trends were described. After addressing multicollinearity and spatial autocorrelation, a geographically and temporally weighted regression (GTWR) model was applied to assess the heterogeneous impacts of economic, transportation, healthcare, and natural factors (2018-2020). Results　From 2004 to 2023, the age-standardized incidence rate (EAPC=5.51) increased, while age-standardized mortality (EAPC=-5.16) declined. Dengue incidence rate rose nationwide, peaking in September. The GTWR model demonstrated strong explanatory power (R2=0.739, adjusted R2=0.719, AICc=340.492). Based on the spatiotemporal distribution of fitted coefficients, per capita disposable income and passenger turnover showed negative correlations with dengue incidence across provinces. International tourist arrivals and average temperature exhibited positive nationwide correlations. Health personnel per 1,000 population correlated positively only in eastern developed provinces. Average precipitation correlated positively north of the Qinling-Huaihe Line but negatively south of it. Conclusion　The incidence of Dengue in most provinces and cities of our country has been on the rise from 2004 to 2020, with the peak of the disease occurring in September. These determinants significantly influence dengue transmission but exhibit marked spatial heterogeneity, necessitating region-specific prevention and control measures.

Key words: Dengue, Incidence, Estimated annual percentage change, Spatiotemporal analysis, Geographically and temporally weighted regression, Influencing factors

摘要： 背景　登革热是我国蚊传疾病中发病率上升较快的媒介传染病之一，流行范围在近几年迅速向北部地区蔓延。目的　分析 2004—2023 年中国登革热的疾病负担、时空分布特征及其影响因素。方法　利用 2025 年全球疾病负担（GBD）数据库、中国公共卫生科学数据中心、《中国卫生健康统计年鉴》、国家数据及天气后报网数据库，分析中国登革热发病及死亡的年百分比变化（EAPC），描述性分析登革热发病率的时间地域分布及流行趋势，最后通过多重共线性及空间自相关检验后，利用时空地理加权回归（GTWR）模型结合 2018—2020年经济、交通、医疗卫生及自然4个维度的影响因素进行省级地域异质性分析。结果　2004—2023年中国登革热年龄标化发病率呈上升趋势（EAPC=5.51），年龄标化死亡率呈下降趋势（EAPC=-5.16）。我国大部分省市登革热发病率在 2004—2020 年间呈上升趋势，发病高峰为 9 月份。依据 2018—2020 年中国登革热发病率数据构建的 GTWR 模型，R2=0.739，Adjusted R2=0.719，AICc=340.492，能够较好地解释自变量对于登革热发病的影响。各变量拟合系数时空分布图显示全体居民人均可支配收入（元）及旅客周转量（亿人公里）在各省市与登革热发病率呈负相关关系，接待国际游客（百万人次）及平均气温在各省市与登革热发病率呈正相关关系，每千人口卫生技术人员数仅在东部经济发达省份与登革热发病率呈正相关关系，平均降水量（mm）在我国秦岭淮河以北主要与登革热发病率呈正相关关系，在秦岭淮河以南主要与登革热发病率呈负相关关系。结论　我国大部分省市登革热发病率在 2004—2020年间呈上升趋势，发病高峰为9月份。全体居民人均可支配收入（元）、及旅客周转量（亿人公里）、接待国际游客（百万人次）、每千人口卫生技术人员数、平均气温及平均降水量（mm）均对我国登革热的发病具有显著影响，但存在较强的地域异质性，因此应针对各地的具体情况制定针对性的防控措施。

关键词: 登革热, 发病率, 估计的年百分比变化, 时空分析, 时空地理加权回归, 影响因素