发热呼吸道症候群与任一病原菌检出对应分析

Multiple correspondence analysis on association of pathogenic isolates with incidents of febrile respiratory syndrome based on sentinel surveillance data from western China

目的通过多重对应分析探索发热呼吸道症候群病例特征、标本及其任一病原菌检出的关联关系,为提高哨点医院感染性疾病监测质量提供依据。方法运用SAS 9.1对应分析过程对基于西部4省哨点医院发热呼吸道症候群监测病例资料及其标本实验室检测有关数据进行多重对应分析,并通过变量分组组合对降维效果进行优化、使结果趋于稳定。结果标本来源于3 157例病例,采集标本前3位分别为鼻/咽拭子(39.31%)、痰(31.45%)、血(16.47%)。任一病原菌检测阳性率为15.24%。省级医院、住院病例、采样前1周有抗生素用药经历与任一病原菌检出结果的阴性相关联;市级及以下医院、门急诊病例及未用抗生素与任一病原菌检出结果阳性相关联。14岁以下儿童病例、鼻/咽拭子标本与任一病原菌检出结果阴性关联;痰标本与任一病原菌检出结果阳性关联。结论西部地区哨点监测发热呼吸道症候群病例的病原菌检出率低,应扩大哨点医院的覆盖效率、加强省级医院门诊病例的监测。

Objective To explore the association of pathogenic isolates with febrile respiratory syndrome（FRS） inci- dents with multiple correspondence analysis and to provide a basis for improving the quality of infectious disease sentinel surveillance. Methods The data on registered FRS cases and laboratory pathogenic detection for the cases in 4 provinces of western China were collected from sentinel surveillance system and analyzed with SAS 9. 1. Subgroup analysis and combi- nations of variables were applied to improve dimension reduction and result. Results The detection rate of pathogens was 15.24% for a total of 3 157 specimens of nose/throat swab （ 39.31% ）, sputum （ 31.45 % ）, and blood （ 16. 47 % ） collected from the FRS cases. There were a lower pathogen detection rate among the FRS cases from provincial hospitals, hospital- ized,and with antibiotics use a week before the specimen sampling and a higher rate among those from municipal and county- level hospitals,being attendees of outpatient clinic or emergency department, and without antibiotics use. A lower patho- gen detection rate was observed among the FRS cases under the age of 14 years and among the nose/throat swab speci-mens;whereas the pathogen detection rate was higher for sputum specimens. Conclusion The pathogen detection rate was low among FRS incidents registered by sentinel surveillance in western China, suggesting that the scope of FRS senti- nel surveillance should be enlarged and the surveillance in outpatient clinic needs to be improved in provincial hospitals.