Modeling seasonal leptospirosis transmission and its association with rainfall and temperature in Thailand using time-series and ARIMAX analyses
Modeling seasonal leptospirosis transmission and its association with rainfall and temperature in Thailand using time-series and ARIMAX analyses
复制标题
DOI：
10.1016/s1995-7645(12)60095-9
复制
发表时间：
2012-07-01
影响因子：
3.1
通讯作者：
Triampo, Wannapong
中科院分区：
文献类型：
--
作者：
Chadsuthi, Sudarat;Modchang, Charin;Triampo, Wannapong
目的：研究钩端螺旋体病发病数与季节的关系，以及与气候因素的关系。方法：采用时间序列分析方法，研究钩端螺旋体病发病数的时间变化。采用自回归滑动平均模型（ARIMA）对数据进行曲线拟合，预测下一个钩体病病例。结果如下：我们发现，降雨量与钩端螺旋体病病例在这两个地区的利益，即泰国的北方和东北部地区，而温度发挥了作用，只有在东北部地区。多变量ARIMA（ARIMAX）模型的使用表明，降雨因素（8个月的滞后）产生的最佳模型为北方地区，而模具，降雨因素（10个月的滞后）和温度（8个月的滞后）是该地区最好的。结论：该模型能够反映两个地区钩端螺旋体病的发病趋势，并能很好地反映两个地区的实际情况。这些模型也可以用来相当准确地预测下一个季节性高峰。
Objective: To study the number of leptospirosis cases in relations to the seasonal pattern, and its association with climate factors. Methods: Time series analysis was used to study the time variations in the number of leptospirosis cases. The Autoregressive Integrated Moving Average (ARIMA) model was used in data curve fitting and predicting the next leptospirosis cases. Results: We found that the amount of rainfall was correlated to leptospirosis cases in both regions of interest, namely the northern and northeastern region of Thailand, while, the temperature played a role in the northeastern region only. The use of multivariate ARIMA (ARIMAX) model showed that factoring in rainfall (with an 8 months lag) yields the best model for the northern region while the mould, which factors in rainfall (with a 10 months lag) and temperature (with an 8 months lag) was the best for the region. Conclusions: The models are able to show the trend in leptospirosis eases and closely lit the recorded data in both regions. The models can also be used to predict the next seasonal peak quite accurately.