Objective: To use the spatiotemporal distribution model and INLA algorithm to study the spatiotemporal characteristics and influencing factors of tuberculosis in Shanghai and to provide a theoretical basis for formulating regional tuberculosis epidemic prevention and control measures. Methods: Based on the data of registered pulmonary tuberculosis cases in Shanghai during 2013-2020 derived from the tuberculosis management information system of China Disease Control and Prevention Information System, the hierarchical Bayesian model was adopted to fit the tuberculosis case data, identify the spatiotemporal variation characteristics of tuberculosis, and explore the potential socioeconomic characteristics and other factors related to health services and spatiotemporal characteristics. Results: From 2013 to 2020, 29 281 registered tuberculosis cases were reported in Shanghai, with an average annual incidence of 25.224/100 000. From 2013 to 2020, the incidence trend increased first and then decreased, the highest incidence was reported in 2014 (27.991/100 000). The incidence of tuberculosis in Shanghai is characterized by spatial clustering. Through the spatial characteristics and risk analysis of the reported incidence of tuberculosis, it is found that the high-risk area of tuberculosis in Shanghai is the suburban communities, whereas downtown communities are the low-risk areas. The incidence risk of pulmonary tuberculosis is associated with the gross domestic product per capita (RR=0.48), the number of beds per 10 000 persons (RR=0.56), the normalized vegetation index (RR=0.50), and the night light index (RR=0.80). Conclusions: With the steady progress of tuberculosis prevention and control in the central urban area of Shanghai, special attention should be paid to the prevention and control in the suburbs further to improve the social and economic level in the suburbs and increase the coverage rate of urban green space, to reduce the incidence of tuberculosis and reduce the disease burden of tuberculosis in Shanghai.
目的： 利用时空分布模型和集成嵌套拉普拉斯近似算法研究上海市肺结核的时空特征及其影响因素，为区域化肺结核疫情防控措施的制定提供理论依据。 方法： 基于中国疾病预防控制信息系统结核病管理信息系统下载的2013-2020年上海市户籍肺结核病例，采用层次贝叶斯模型对肺结核病例数据进行拟合，识别肺结核的时空变化特征，从时空角度，评估肺结核发病风险与人口、经济、卫生服务等因素的关系。 结果： 2013-2020年，上海市共报告肺结核病例29 281例，年均报告发病率为25.224/10万。2013-2020年的发病率呈先上升后下降的趋势，2014年的报告发病率最高（27.991/10万）。上海市肺结核的发生具有空间聚集性，通过肺结核报告发病率的热点分析和风险分析发现，上海市肺结核的高风险区域为郊区，低风险区域为中心城区。肺结核的发病风险与地区人均生产总值（RR=0.48）、每万人医疗机构床位数（RR=0.56）、归一化植被指数（RR=0.50）、夜间灯光指数（RR=0.80）有关。 结论： 在上海市中心城区肺结核防控工作的稳步推进下，需要格外关注郊区的防控。在后续的研究中，应进一步探索肺结核发病与归一化植被指数和夜间灯光指数之间的联系。.