文章摘要
王震坤,陈知水,杜艾桦,王从义,刘虹,王子伟,胡继发.基于大数据与有效距离模型的突发急性传染病宏观预警及防控管理工作探讨:COVID-19疫情数据的启示[J].中华流行病学杂志,,():
基于大数据与有效距离模型的突发急性传染病宏观预警及防控管理工作探讨:COVID-19疫情数据的启示
Discussion on early warning, prevention and control of emerging infectious diseases from a macroscopic perspective based on big data and effective distance model: enlightenment of COVID-19 epidemic data in China
收稿日期:2020-03-06  出版日期:2020-04-27
DOI:10.3760/cma.j.cn112338-20200306-00269
中文关键词: 新型冠状病毒肺炎;有效距离;人口迁徙;传染病;预警防控
英文关键词: COVID-19;Effective distance;Population floating;Infectious diseases;Early warning, prevention and control
基金项目:国家自然科学基金(81903396)
作者单位E-mail
王震坤 华中科技大学同济医学院附属同济医院科研处, 武汉 430030  
陈知水 华中科技大学同济医学院附属同济医院科研处, 武汉 430030  
杜艾桦 华中科技大学同济医学院附属同济医院科研处, 武汉 430030  
王从义 华中科技大学同济医学院附属同济医院科研处, 武汉 430030  
刘虹 华中科技大学同济医学院附属同济医院科研处, 武汉 430030  
王子伟 华中科技大学同济医学院附属同济医院科研处, 武汉 430030  
胡继发 华中科技大学同济医学院附属同济医院科研处, 武汉 430030 jfahu@tjh.tjmu.edu.cn 
摘要点击次数: 1030
全文下载次数: 631
中文摘要:
      目的 从宏观视角利用新型冠状病毒肺炎(COVID-19)疫情数据,分析有效距离与疫情传播轨迹、时间和规模之间的关系,为今后疫情防控提供科学依据。方法 收集整理截至2020年2月23日我国各地COVID-19首例确诊患者的住院治疗/隔离治疗日期以及累计确诊病例数,利用“百度迁徙”基于地理位置的服务大数据平台(LBS)获取武汉市到各地的迁出人口比例数据,建立有效距离模型和线性回归模型,从省级和市级层面分别对有效距离与疫情抵达时间及累计确诊病例级数的关系进行分析。结果 不论在省级层面还是市级层面上,武汉市到目的地的有效距离与COVID-19疫情抵达时间及累计确诊病例级数都存在明显的线性关联,各线性模型回归系数均有统计学意义(P<0.001)。在省级层面上,有效距离可以解释其与抵达时间模型71%的变异,解释其与累计确诊病例级数模型90%的变异;在市级层面上,有效距离可以解释其与抵达时间模型66%的变异,解释其与累计确诊病例级数模型85%的变异。结论 模型拟合程度较好,LBS大数据与有效距离模型能够用于对疫情传播轨迹、时间和规模等进行估计,为突发急性传染病宏观预警及防控管理工作提供有益参考。
英文摘要:
      Objective To provide a system for warning, preventing and controlling emerging infectious diseases from a macroscopic perspective, using the COVID-19 epidemic data and effective distance model. Methods The dates of hospitalization/isolation treatment of the first confirmed cases of COVID-19 and the cumulative numbers of confirmed cases in different provinces in China reported as of 23 February, 2020 were collected. The Location Based Service (LBS) big data platform of "Baidu Migration" was employed to obtain the data of the proportion of the floating population from Wuhan to all parts of the country. Effective distance models and linear regression models were established to analyze the relationship between the effective distance and the arrival time of the epidemic as well as the number of cumulative confirmed cases at provincial and municipal levels. Results The arrival time of the epidemic and the cumulative number of confirmed cases of COVID-19 had significant linear relationship at both provincial and municipal levels in China, and the regression coefficients of each linear model were significant (P<0.001). At the provincial level, the effective distance could explain about 71% of the variation of the model with arrival time along with around 90% of the variation for the model in the cumulative confirmed case magnitude; at the municipal level, the effective distance could explain about 66% of the variation for the model in arrival time, and about 85% of the variation of the model with the cumulative confirmed case magnitude. Conclusions The fitting degree of the models are good. The LBS big data and effective distance model can be used to estimate the track, time and extent of epidemic spread to provide useful reference for early warning, prevention and control of emerging infectious diseases.
查看全文   Html全文     查看/发表评论  下载PDF阅读器
关闭