文章摘要
康乐妮,张韶凯,赵方辉,乔友林.逆概率权重法在诊断试验评价证实偏倚中的应用[J].中华流行病学杂志,2014,35(3):329-332
逆概率权重法在诊断试验评价证实偏倚中的应用
Implication of inverse-probability weighring method in the evaluation of diagnostic test with verification bias
收稿日期:2013-09-29  出版日期:2014-09-18
DOI:
中文关键词: 逆概率权重法;证实偏倚;灵敏度;特异度;最大似然估计
英文关键词: Ehrlichia bacteria;Ehrlichiosis
基金项目:国家自然科学基金
作者单位E-mail
康乐妮  qiaoy@cicams.ac.cn 
张韶凯   
赵方辉 
北京协和医学院中目医学科学院肿瘤医院肿瘤研究所流行病室 100021
 
 
乔友林 
北京协和医学院中目医学科学院肿瘤医院肿瘤研究所流行病室 100021
 
 
摘要点击次数: 1729
全文下载次数: 959
中文摘要:
      如何估计和校正筛查或诊断试验中存在的证实偏倚,文中通过宫颈癌筛查实例,采用逆概率权重法R软件CompareTesl:s校诈其灵敏度和特异度,利用随机抽样方法生成新数据,将逆概率权重法计算的灵敏度和特异度,与传统计算方法以及最大似然估计方法计算得到的灵敏度和特异度进行比较。结果表明HPV自检法的真实灵敏度和特异度分别为83.53%(95%CI:74.23~89.93)和85.86%(95%C1:84.23~87.36)。随机抽样结果显示,传统方法计算的灵敏度和特异度分别为90.48%(95%C1:80.74-95.56)和71.96%(95%C1:68.71~75.00),采用逆概率权重法校正后的灵敏度和特异度分别为82,25%(95%CI:63。11-92.62)和85。80%(95%C1:85。09~86.47);采用最大似然估计法校正后的灵敏度和特异度分别为80.13%(95%CI:66.81.93.46)和85.80%(95%CI:84.20~87.41)。表明在复杂抽样情况下,逆概率权重法能够有效校正存在证实偏倚的灵敏度和特异度。
英文摘要:
      ObjectiveTo evaluate and adjust the verifcation bias existed in the screening ordiagnostic tests.Inverse-probability weighting method was used to adjust the sensitivity and specificityof the diagnostic tests.with an example of cervical cancer screening used to introduce the CompareTsts package in R software which could be implemented.Sensitivity and specificity calculated fromthe traditional method and maximum likelihood estimation method were compared to the Results fromInverse.probability weighting method in the random.sampled example.The true sensitivity andspecificity ofthe HPV self-sampling test were 83.53%(95%CI:74.23-89.93)and 85.86%(95%6"/:84.23-87.36).In the analysis of data with randorely missing vefification by gold standard.thesensitivity and specificity calculated by traditional method were 90.48%(95%CI:80.74-95.56)and71.96%(95%C7:68.71-75.00),respectively.Thc adjusted sensitivity and specificity underthe use ofInverse-probability weighting method were 82.25%(95%CI:63.11-92.62)and 85.80%(95%CI: 85.09-86.47).respectively.whereas they were 80.13%(95%C,:66.81-93.46)and 85.80%(95%CI:84.20-87.41)under the maximum likelihood estimation method.The inverse-probability weightingmethod could effectively adiust the,.;ensitivity and specificity of a diagnostic test whenverification bias existed,especially when complex sampling appeared.
查看全文   Html全文     查看/发表评论  下载PDF阅读器
关闭