Influence of COVID-19 prevention and control measures on the temporal and spatial evolution of epidemic situation

Objective To understand the spatiotemporal distribution and spatial clustering of coronavirus disease 2019 (COVID-19), and to evaluate the impact of comprehensive prevention and control measures on the COVID-19 epidemic. Methods Three periods was divided according to the measures of first level emergency response and the resumption of work and production. Data of diagnosed number in all provinces and municipalities was collected, and the diagnosed number per week in each time period was calculated. The spatial autocorrelation method was used to analyze the temporal and spatial evolution of COVID-19 epidemic. Results In the first stage, a significant positive spatial correlation was observed rounding Hubei province (Morans'I=0.144, Z =2.517, P =0.016). Local spatial autocorrelation analysis showed a high-high clustering between Hubei and its adjacent provinces, and a low-low clustering in these provinces far away from Hubei. After the measures of first level emergency response being taken(the second stage), the COVID-19 epidemic presented a random distribution (Morans'I=-0.017, Z =0.597, P =0.251). A high-low clustering between Hubei and its adjacent provinces, and a low-low clustering in these provinces far away from Hubei was observed. Since the time of work and production was resumed (the third stage), the COVID-19 epidemic remained a random distribution (Morans ' I=-0.055, Z =-0.857, P =0.131). A high-low clustering between Guangdong and Jiangxi, and a low-low clustering in Sichuan was observed. Conclusion The comprehensive prevention and control measures adopted in China have effectively controlled the spread of COVID-19 epidemic throughout the country. At present, the epidemic posed a sporadic state. A closely monitoring and comprehensive prevention and control measures should be maintained to prevent the rebound of COVID-19 epidemic. 摘要：目的 分析新型冠状病毒肺炎(COVID-19)在不同时期的地理空间分布、空间聚集性特征, 评价综合防控措 施对疫情发展的影响。 方法 根据我国各地区采取I级响应措施及复工复产时间将疫情流行分为三个时间段。收集 我国内地各省、直辖市COVID-19动态确诊数据, 计算每个时间段的周确诊病例数;运用空间自相关法对COVID-19确 诊情况进行统计分析。 结果 在第一阶段, 疫情呈现以湖北为中心的典型空间正相关(Morans’I=0.144, Z=2.517, P =0.016);局部聚集表现为湖北与周边省份的髙-髙型和与湖北相距较远省份的低-低型分布。自各地区启动I级响应措 施以后, 疫情在第二阶段呈现随机分布(Morans’I=-0.017, Z =0.597, P =0.251), 局部聚集表现为湖北与周边省份的髙-低 型和与湖北相距较远省份的低-低型分布。自各地区复工复产后, 疫情在第三阶段仍呈现随机分布(Morans’ I=-0.055, Z =-0.857, P =0.131);局部聚集表现为广东与江西的髙-低型及四川地区的低-低型分布。 结论 我国采取的综合防控 措施有效地控制了全国疫情蔓延, 目前疫情总体处于散发流行状态 , 应防止疫情反弹