2019-2023年珠海市香洲区初治痰涂阳肺结核患者影响因素的Logistic回归模型  

Influencing Factors of Patients with Primary Sputum Smear Positive Tuberculosis in Xiangzhou District,Zhuhai City,2019-2023:A Logistic Regression Model

作　　者：江澄瑜 张定梅[1] 黄雅怡 罗小青[2] 李文玲 JIANG Cheng-yu;ZHANG Ding-mei;HUANG Ya-yi;LUO Xiao-qing;LI Wen-ling(School of Public Health,Sun Yat-sen University,Guangzhou,Guangdong,510080,China;Infectious Disease Department,The Fifh Affiliated Hospital of Sun Yat sen University,Zhuhai,Guangdong,519000,China)

机构地区：[1]中山大学公共卫生学院,广东广州510080 [2]中山大学附属第五医院,广东珠海519000

出　　处：《现代生物医学进展》2025年第6期1073-1085,共13页Progress in Modern Biomedicine

摘　　要：目的:调查2019-2023年珠海市香洲区初治痰涂阳肺结核患者的流行病学特征,及构建其影响因素的Logistic回归预测模型,为该地的结核病防治策略的制定提供参考。方法:通过"中国疾病预防控制信息系统"收集资料,应用描述性研究及Logistic回归分析方法,对患者特征及影响强化期痰转阴的因素进行分析,构建Logistic回归预测模型,并分析其预测效能。结果:筛选出初治痰涂阳肺结核患者1005例,强化期痰转阴率为91%,男女性别构成比为1.8:1,45-64岁年龄段、家务及待业职业、省间流动、追踪发现、夏季6月份发病的患者占比最多。多因素Logistic回归分析显示,性别为男、45-64岁年龄段、春季发病是强化期痰未转阴的危险因素。对应的回归预测模型经受试者工作特征曲线(receiver operating characteristic curve,ROC)分析,结果显示该预测模型的受试者工作特征曲线下面积(area under the receiver operating characteristic curve,ROC-AUC)及95%置信区间(95%confidence interval,95%CI)为0.728(0.631~0.825),预测敏感度为82%,准确度为67%。结论:珠海市香洲区应重点关注男性、青中年、省间流动、家务及待业人群、夏季尤其是6月份的肺结核发病群体,男性、青中年、春季发病增加了强化期痰涂片转阴失败的风险,通过影响因素构建的预测模型对肺结核痰转阴有较好的预测效能。应加强对重点人群和时期的健康宣传和教育,制定针对性的结核病防控措施,以促进痰菌转阴,减少疾病的传播。Objective:To investigate the epidemiological characteristics of patients with primary sputum smear positive tuberculosis in Xiangzhou District,Zhuhai city from 2019 to 2023,and construct a Logistic regression model of risk factors to provide references for the formulation of tuberculosis control strategies in this area.Methods:Data was collected through the'China Disease Prevention and Control Information System',descriptive research and Logistic regression analysis methods were applied to analyze patient characteristics and factors affecting sputum conversion during the intensive phase of treatment.A Logistic regression predictive model was constructed and its predictive efficacy was analyzed.Results:A total of 1005 newly diagnosed sputum-positive pulmonary tuberculosis patients were selected,with a sputum conversion rate of 91%during the intensive phase.The gender ratio was 1.8:1,with the highest proportion of patients in the 45-64 age group,engaged in housework or unemployed,inter-provincial mobility,traced and identified,and with onset in June during the summer season.Multivariate Logistic regression analysis indicated that being male,aged 45-64,and having onset in spring were risk factors for not achieving sputum conversion during the intensive phase.The corresponding regression predictive model,after ROC analysis,showed that the model had a ROC-AUC(95%CI)of 0.728(0.631-0.825),with a predictive sensitivity of 82%and an accuracy of 67%.Conclusion:In Zhuhai's Xiangzhou District,special attention should be given to male,middle-aged,inter-provincial mobile,housework,and unemployed populations,especially the tuberculosis patient group that develops in June during the summer.Being male,middle-aged,and having onset in spring increases the risk of failure to convert sputum smears during the intensive phase.The predictive model built from influencing factors has good predictive efficacy for sputum conversion in tuberculosis.Strengthening health promotion and education for key groups and periods,and formulating target

关 键 词：初治肺结核 流行病学特征 强化期痰转阴 影响因素 预测模型 

分 类 号：R3[医药卫生—基础医学] R181.3R521