Abstract: Objective To explore the spatial and temporal distribution characteristics of human brucellosis epidemic in Hubei province, China, and to predict the incidence trend. Methods The spatial and temporal distribution of human brucellosis epidemic in Hubei province from 2007 to 2017 was investigated by traditional epidemiology, spatial epidemiology, and time series analysis, combined with SaTScan 9.4, ArcGIS 10.0, and SPSS 18.0 softwares. Results From 2010 to 2017, there were five areas with spatial-temporal clustering of brucellosis incidence in Hubei province; all cases occurred in 2015-2017. There was one class Ⅰ cluster area covering eight counties (districts) with a maximum scanning radius of 105.94 km, which was mainly distributed in Suizhou, Xiangyang, and Jingmen; the clustering time was April to July, 2015, and 113 cases were reported. There were four class Ⅱ cluster areas, mainly distributed in five cities (prefectures), i.e., Huanggang, Huangshi, Xianning, Wuhan, and Enshi Tujia and Miao Autonomous Prefecture, and covering nine counties (districts) in the southeast and west of Hubei province. The simple seasonal model based on the exponential smoothing method had a good fitting effect, and the incidence of brucellosis in Hubei province from January to June, 2018 was predicted to be slightly lower than that in the same period in 2016-2017. Conclusion The epidemic situation of brucellosis in Hubei province has entered into a rapid "rising period" from the lower "plateau stage". The brucellosis epidemic has spread to a wider area, and the epidemic situation tends to be complex, with some areas showing spatial- temporal clustering. With the simple seasonal model fitting, the predictive model of brucellosis incidence trend in Hubei province has a high prediction accuracy, which can be used to strengthen surveillance and early warning, and promptly adjust and improve the prevention and control strategy of brucellosis.

Key words: Brucellosis, Spatial and temporal distribution, Trend prediction

摘要： 目的 探索湖北省布鲁氏菌病（布病）人间疫情时空分布特征，进行发病趋势预测。方法 综合应用传统流行病学、空间流行病学、时间序列分析等方法，结合SaTScan 9.4、ArcGIS 10.0和SPSS 18.0等软件，探讨2007－2017年湖北省人间布病疫情发病时空分布规律。结果 2010－2017年湖北省布病发病存在时空聚集性的区域有5个，均发生在2015－2017年。一类聚集区有1个，区域覆盖8个县（区），最大扫描半径为105.94 km，主要分布在随州、襄阳、荆门3个市，聚集时间为2015年4－7月，报告病例数113例。二类聚集区4个，主要分布在黄冈、黄石、咸宁、武汉、恩施土家族苗族自治州5个市（州），覆盖湖北省东南部和西部9个县（区）；使用指数平滑法中的简单季节性模型拟合效果好，预测2018年1－6月湖北省布病发病数与2016－2017年同期比略降。结论 湖北省布病疫情由较低“平台期”进入较快“上升期”，布病疫情波及地区扩大，疫情形势趋于复杂，且部分地区存在时间和空间的聚集现象。运用简单季节性模型拟合，湖北省布病发病趋势预测模型有较高的预测精度，可结合预测模型，加强监测预警，及时调整和完善布病防控策略。

关键词: 布鲁氏菌病, 时空分布, 趋势预测