论文部分内容阅读
目的研究气象条件的变化对甲型H1N1流感活动强度的影响,建立气象因子对甲型H1N1流感阳性检出率的预报模型。方法收集浙江省2009年4月—2011年1月甲型H1N1流感疫情资料、各哨点医院的监测资料以及同期气象资料,在采用Spearman相关分析甲型H1N1流感活动强度与有关气象因子的相关关系的基础上,通过卡方自动交互检测方法建立决策树预报模型。结果甲型H1N1流感阳性检出率与周平均气压(r=0.50)、最高气压(r=0.51)、降水量(r=-0.23)、最低气压(r=0.50)、平均气温(r=-0.40)、最高气温(r=-0.41)、最低气温(r=-0.39)、平均风速(r=-0.22)、气温周较差(r=-0.30)和周平均温差(r=-0.30)等气象因子存在相关性(P<0.05)。影响甲型H1N1流感阳性检出率的主要气象因子为最低气压、平均风速和降水量(P均<0.05)。将甲型H1N1流感阳性检出率分为不同等级与气象因子建立预报模型,模型预测的正确率为66.67%。结论最低气压、平均风速、降水量等气象因子与甲型H1N1流感活动强度密切相关,可利用决策树建立模型作预测预报。
Objective To study the influence of the change of meteorological conditions on the activity of Influenza A (H1N1) and establish the forecasting model of the positive detection rate of Influenza A (H1N1) by meteorological factors. Methods The epidemic data of Influenza A (H1N1), the monitoring data of sentinel hospitals and the meteorological data of the same period were collected from April 2009 to January 2011 in Zhejiang Province. Spearman correlation analysis was used to analyze the correlation between the activity intensity of Influenza A (H1N1) and related meteorological factors Based on the above, a decision tree prediction model is established by means of automatic interaction detection of chi-square. Results The positive rate of detection of Influenza A (H1N1) was positively correlated with the average weekly air pressure (r = 0.50), maximum air pressure (r = 0.51), precipitation (r = -0.23), minimum air pressure 0.40), the highest temperature (r = -0.41), the lowest temperature (r = -0.39), the average wind speed (r = -0.22), the poor temperature week (r = -0.30) and the weekly average temperature difference Other weather factors such as correlation (P <0.05). The main meteorological factors influencing the positive detection rate of Influenza A (H1N1) were the minimum pressure, average wind speed and precipitation (P <0.05). The detection rate of influenza A (H1N1) was divided into different grades and meteorological factors to establish the prediction model. The accuracy of the model was 66.67%. Conclusions The meteorological factors such as minimum pressure, average wind speed and precipitation are closely related to the activity intensity of Influenza A (H1N1). A decision tree can be used to establish a model for forecasting.