基于时序特征的手足口病疫情模式聚类研究

Cluster analysis of epidemic characteristics of hand-foot-and-mouth disease based on temporal characteristics

摘要:

目的：根据全国31个地区2010至2015年手足口病发病的时间序列特征，划分不同的疫情模式，为该疾病防控的统一规划和实施提供科学依据。方法：提取各地区手足口病发病率的时间序列特征，包括季节波动、趋势、自相关和混沌等；通过层次聚类分析，划分不同的时间序列疫情模式。结果：聚类分析共产生3类疫情模式，分别为非季节模式、单发季节模式和多发季节模式。结论：对非季节模式地区，需随时加强疾病监控，进一步分析其相关因素；对单发季节模式和多发季节模式，仅需在周期性高发来临时作好相应的预防措施，从而降低疾病大面积流行的可能性。

Abstract:

Objective：to provide a scientific basis for the prevention and control of the hand-foot-and-mouth disease（HFMD） by di－viding different epidemic models according to the time series characteristics in 31 regions from 2010 to 2015. Methods：Time sequence features of HFMD incidence in different regions was extracted，including seasonal fluctuations，trends，autocorrelation and chaos，etc. Different time series epidemic patterns were divided by hierarchical clustering analysis. Results：Cluster analysis produced three types of epidemic patterns，namely，non-seasonal mode，single-season and multi-season seasonal pattern. Conclusion：For non-seasonal patterns，disease surveillance should be strengthened at any time to further analyzing the relevant factors. In the single-season and multi-season model，the appropriate preventive measures should be taken only in the high incidence period to reduce the possibility of a large area of epidemic disease.