基于灵敏度和特异度任意赋权的广义标准化诊断符合率  

Generalized Standardized Diagnostic Efficiency Rate Based on Weighted Sensitivity and Specificity

作　　者：李丹玲[1] 陈平雁[1] 

机构地区：[1]南方医科大学公共卫生与热带医学学院生物统计学系,510515

出　　处：《中国卫生统计》2012年第6期794-796,800,共4页Chinese Journal of Health Statistics

基　　金：国家自然科学基金资助(30972554);广东省自然科学基金资助(915180200400001)

摘　　要：目的建立一种基于灵敏度(SEN)和特异度(SPE)任意赋权的广义标准化诊断符合率统计方法。方法广义标准化诊断符合率(e')的构造满足以下两个原则:(1)灵敏度和特异度的权重(w)之和为1;(2)满足特殊性:当灵敏度和特异度等权时,广义标准化诊断符合率等于标准化诊断符合率(e),即有e'=e。所构造广义标准化诊断符合率为:e1'=[(n1+m1)·2w·SEN1+(m2+n2)·2(1-w)·SPE1]/(n+m),e2'=[(n1+m1)·2w·SEN2+(m2+n2)·2(1-w)·SPE]/(2n+m),(0≤w≤1)。根据中心极限定理,推导出e'的标准误和两个e'比较的Z统计推断方法,进一步推导出权重w的变化对检验统计量Z的影响。结果所构造的广义标准化诊断符合率满足上述构造两个原则。结论本研究所建立的广义标准化诊断符合率方法解决了应用中对灵敏度和特异度有不同赋权要求的问题,为诊断试验评价提供了新的手段。Objective To develop a generalized standardized diagnostic efficiency rate（e＇）method for evaluating and comparing diagnostic tests based on weighted sensitivity and specificity.Methods Two principles of constructing e＇ are as follows：firstly,the sum of two weights which are attached to the sensitivity and specificity should equal to 1;secondly,generalized standardized diagnostic efficiency rate equals to standardized diagnostic efficiency rate when the sensitivity and specificity have the same weights,that is e＇=e.Then,the e＇ is defined by e1＇=[（n1＋m1）·2w·SEN1＋（m2＋n2）·2（1-w）·SPE1]/（n＋m）,e2＇=[（n1＋m1）·2w·SEN2＋（m2＋n2）·2（1-w）·SPE2]/（n＋m）（0≤w≤1）.Methods According to the central-limit theorem,we obtain the standard error of e＇,and propose a statistical inference method to compare two generalized standardized diagnostic efficiency rate.Furthermore,we also deduce the test statistics Z can be either a monotonously increasing/decreasing function or non-monotone function of the weight w under different conditions.Results The proposed e＇ satisfied the above-mentioned two principles.Conclusion For different weights attached to the sensitivity and specificity,e＇ can be used to deal with such kinds of problems as provide a new statistical tool to evaluate diagnostic tests.

关 键 词：标准化诊断符合率 权重 灵敏度 特异度 

分 类 号：R195[医药卫生—卫生统计学]