构建及评价公共场所喷泉水嗜肺军团菌污染影响因素的logistic回归模型

发布时间：2018-08-12 19:23

【摘要】：[目的]探讨公共场所喷泉水嗜肺军团菌污染的影响因素,拟合logistic回归方程用于喷泉水嗜肺军团菌污染的预测,为预防和控制喷泉水军团菌污染提供科学依据。[方法]选择深圳市公共场所正常运行的80座喷泉作为研究对象,采用问卷调查、现场监测及实验室检测等方式收集相关数据。2015年7—12月收集50座喷泉(其中包括30座住宅小区喷泉和20座公共建筑周围的喷泉)相关资料,利用logistic回归法对可能影响喷泉水嗜肺军团菌污染的因素进行分析并建立回归模型,绘制受试者工作特征(ROC)曲线,确定喷泉水嗜肺军团菌污染的回归模型的最佳临界值。采用2016年1—7月收集的30座喷泉(其中包括20座住宅小区喷泉和10座公共建筑周围的喷泉)相关数据验证上述模型的准确率。[结果]单因素logistic回归分析发现,游离性余氯含量升高是嗜肺军团菌污染的保护因素(OR=0.986);运行年限的增加、喷泉水浊度升高、溶解性总固体含量升高是嗜肺军团菌污染的危险因素(OR=1.096、1.261、1.100)。经多因素logistic回归分析筛选出游离性余氯(OR=0.952,P0.05)、浊度(OR=1.314,P0.05)和溶解性总固体(OR=1.098,P0.05)3个因素用于模型的拟合,其ROC曲线下面积为0.877(95%CI:0.808~0.997),该模型预测准确率为80.0%。[结论]logistic回归模型发现喷泉水游离性余氯、浊度和溶解性总固体是影响嗜肺军团菌污染的主要因素。该模型预测准确率较高,对公共场所喷泉水嗜肺军团菌污染的判断有一定的参考价值。
[Abstract]:[objective] to explore the influencing factors of Legionella pneumophila contamination in fountain water in public places, and to apply the fitting logistic regression equation to predict the contamination of Legionella pneumophila in fountain water, so as to provide scientific basis for preventing and controlling Legionella pneumophila pollution in fountain water. [methods] A questionnaire survey was conducted on 80 fountains in public places in Shenzhen. Data are collected by means of on-site monitoring and laboratory testing. In July-December 2015, 50 fountains (including 30 fountains in residential areas and 20 fountains around public buildings) were collected. The factors that may affect the contamination of Legionella pneumophila in fountain water were analyzed by using logistic regression method and the regression model was established. The (ROC) curve of operating characteristics of the subjects was drawn and the optimal critical value of the regression model for the contamination of Legionella pneumophila in fountain water was determined. The accuracy of the model was validated by data from 30 fountains collected between January and July 2016, including 20 fountains in residential areas and 10 around public buildings. [results] by univariate logistic regression analysis, it was found that the increase of free residual chlorine content was the protective factor of Legionella pneumophila (OR=0.986), and the turbidity of fountain water increased with the increase of running life. The increase of soluble total solid content was the risk factor of Legionella pneumophila contamination (OR1. 096 ~ 1. 261 ~ 1. 100). By multivariate logistic regression analysis, three factors were selected to fit the model: free residual chlorine (ORG) 0.952 (P0.05), turbidity (OR1. 314) and dissolved solids (OR1. 098%). The area under the ROC curve was 0.877 (95%CI:0.808~0.997), and the prediction accuracy of the model was 80. 0B. [conclusion] logistic regression model showed that the free residual chlorine, turbidity and soluble solids in fountain water were the main factors affecting the contamination of Legionella pneumophila. The prediction accuracy of the model is high, and it has certain reference value for judging the contamination of Legionella pneumophila in fountain water in public places.
【作者单位】：深圳市南山区疾病预防控制中心环境卫生科;
【基金】：深圳市南山区科技创新局项目(编号:2015071)
【分类号】：R126.4

【参考文献】