Malaria Department, National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention, Zhengzhou, People's Republic of China.
Am J Trop Med Hyg. 2011 Sep;85(3):560-7. doi: 10.4269/ajtmh.2011.11-0156.
Despite significant reductions in the overall burden of malaria in the 20th century, this disease still represents a significant public health problem in China, especially in central areas. Understanding the spatio-temporal distribution of malaria is essential in the planning and implementing of effective control measures. In this study, normalized meteorological factors were incorporated in spatio-temporal models. Seven models were established in WinBUGS software by using Bayesian hierarchical models and Markov Chain Monte Carlo methods. M₁, M₂, and M₃ modeled separate meteorological factors, and M₃, which modeled rainfall performed better than M₁ and M₂, which modeled average temperature and relative humidity, respectively. M₇ was the best fitting models on the basis of based on deviance information criterion and predicting errors. The results showed that the way rainfall influencing malaria incidence was different from other factors, which could be interpreted as rainfall having a greater influence than other factors.
尽管 20 世纪疟疾的整体负担显著减少，但这种疾病在中国仍然是一个重大的公共卫生问题，特别是在中部地区。了解疟疾的时空分布对于规划和实施有效的控制措施至关重要。在这项研究中，归一化气象因素被纳入时空模型。使用贝叶斯层次模型和马尔可夫链蒙特卡罗方法，在 WinBUGS 软件中建立了七个模型。M₁、M₂ 和 M₃ 分别模拟了单独的气象因素，其中模拟降雨量的 M₃ 比分别模拟平均温度和相对湿度的 M₁ 和 M₂ 表现更好。基于偏差信息准则和预测误差，M₇ 是最佳拟合模型。结果表明，降雨量影响疟疾发病率的方式与其他因素不同，这可以解释为降雨量比其他因素的影响更大。
