论文部分内容阅读
目的探讨布雷图指数(BI)、输入病例、气象因子与登革热本地病例发生与否和本地流行严重程度的关系,提出预警界值。方法选择广州市为研究现场,以2002—2013年为研究期限,以月为时间单元,分别选择登革热月本地病例发生与否和月本地感染病例发病数为结局变量,以月均BI、月均输入病例数、月均气温、月最高气温、月最低气温、月均日温差、月总和降雨量、月均降雨量、月均相对湿度和月均气压为自变量,应用分类和回归树(CART)模型对数据进行分析。结果2002—2013年期间共44个月(44/144,30.56%)发生本地感染疫情,共报告病例3 996例,包括3 769例(占94.32%)本地感染病例和227例(占5.68%)输入性病例。本地病例发生与否的分类树模型最终纳入月均输入病例数、月均BI、月最高气温和月最低气温,其中月均输入病例≥3.5例时本地病例发生风险最大,RR值为3.00(2.22~4.05);月均输入病例<3.5例、月均BI≥8.59时RR值为2.40(1.62~3.55);月均输入病例<3.5例、BI<8.59、月最高气温≥31.41℃且月最低气温<24.90℃时RR值为2.18(1.29~3.68)。本地流行严重程度的回归树模型最终纳入月均BI、月均气温和月均输入病例数,其中月均BI≥5.29且月均气温<27.04℃时RR值最大,为5.11(3.36~7.77);月均气温≥27.04℃且月均BI≥9.16时RR值为3.20(2.06~4.96),月均BI<5.29且月均输入病例数≥3.5例时RR值为2.22(1.40~3.53)。结论月均输入病例数和月均BI是本地病例发生与否的2个最重要因素,而月均BI和月均气温则是本地流行疫情严重程度的2个最重要因素。
Objective To explore the relationship between the occurrence of Brett Index (BI), input cases, meteorological factors and the incidence of local dengue cases and the severity of local epidemics and put forward the precautionary boundary value. Methods Guangzhou was selected as the study site. The study period from 2002 to 2013 was taken as the study period. The months were taken as the time unit, and the incidence of local dengue fever and local incidence of infection were selected as the outcome variables. The average monthly BI, monthly average The number of imported cases, monthly average temperature, monthly maximum temperature, monthly minimum temperature, monthly mean daily temperature difference, monthly sum total rainfall, monthly average rainfall, monthly mean relative humidity and monthly mean pressure were used as independent variables. Classification and regression trees (CART The model analyzes the data. Results A total of 44,944 cases (44.54%) of local infections were reported in 44 months (2002-2003). A total of 3 996 cases were reported, including 3 769 cases (94.32%) of local infections and 227 cases (5.68% Enter sex cases. The classification tree model of local cases finally included in the average number of monthly input cases, monthly average BI, monthly maximum temperature and monthly minimum temperature, of which the average monthly incidence of imported cases ≥ 3.5 cases, the local case of the highest risk, RR value of 3.00 (2.22 ~ 4.05). The monthly average input was less than 3.5 cases. The average monthly RR of BI≥8.59 was 2.40 (1.62 ~ 3.55). The average monthly input was less than 3.5 cases, BI was less than 8.59. The monthly maximum air temperature was more than or equal to 31.41 ℃ and the monthly minimum air temperature <24.90 ℃ when the RR value of 2.18 (1.29 ~ 3.68). The regression tree model of local epidemic severity finally included the average monthly BI, monthly mean temperature and monthly average number of imported cases, with the highest monthly RR of 5.11 (3.36-7.77) when the average monthly BI≥5.29 and monthly mean temperature <27.04 ℃. The RR was 2.22 (1.40 ~ 3.53) when the monthly average temperature was more than or equal to 27.04 ℃ and the average monthly BI≥9.16 was 3.20 (2.06 ~ 4.96). The average monthly BI was less than 5.29. Conclusions The monthly average number of imported cases and monthly average BI are the two most important factors in local cases. The monthly average BI and monthly mean temperature are the two most important factors of the severity of the local epidemic.