A non-inferiority test of areas under two parametric ROC curves

被引：12
作者：
Jin, Hua [1 ,2 ]
Lu, Ying [3 ,4 ,5 ]
机构：
[1] S China Normal Univ, Sch Math Sci, Guangzhou 510631, Guangdong, Peoples R China
[2] S China Normal Univ, Ctr Studies Psychol Applicat, Guangzhou 510631, Guangdong, Peoples R China
[3] Univ Calif San Francisco, Dept Radiol & Biomed Imaging, San Francisco, CA 94143 USA
[4] Univ Calif San Francisco, Dept Epidemiol & Biostat, San Francisco, CA 94143 USA
[5] Univ Calif San Francisco, UCSF Helen Diller Family Comprehens Canc Ctr, San Francisco, CA 94143 USA
来源：
CONTEMPORARY CLINICAL TRIALS | 2009年 / 30卷 / 04期
基金：
美国国家卫生研究院;
关键词：
Non-inferiority test; Receiver operating characteristic (ROC) curve; Area under the ROC curve (AUROC); Restricted maximum likelihood (REML); SAMPLE-SIZE; EQUIVALENCE; SENSITIVITIES; TRIALS;
D O I：
10.1016/j.cct.2009.03.003
中图分类号：
R-3 [医学研究方法]; R3 [基础医学];
学科分类号：
1001 ;
摘要：
Equivalence or non-inferiority in diagnostic accuracy of two medical diagnostic tests is a common medical problem. Statistical tests of non-inferiority of diagnostic tests have long been a subject of interest in medicine and biostatistical research. Accuracy of a continuous diagnostic test can be evaluated by the area under a receiver operating characteristic (ROC) curve. A conventional non-inferiority test for areas of two parametric ROC curves has been proposed by Zhou, Obuchowski, and McClish [Zhou XH, Obuchowski NA, McClish DK. Statistical Methods in Diagnostic Medicine. New York: John Wiley: 2002]. In this paper, we showed that this conventional approach is anti-conservative and proposed a new statistical test based on restricted maximum likelihood estimator to achieve designed type I error rate. Simulation studies supported this new test to be appropriate for small sample size or highly correlated diagnostic modalities. (C) 2009 Elsevier Inc. All rights reserved.
引用
收藏
页码：375 / 379
页数：5
相关论文
共 28 条
[1]  
Berger RL, 1996, STAT SCI, V11, P283
[2]   COMPARING THE AREAS UNDER 2 OR MORE CORRELATED RECEIVER OPERATING CHARACTERISTIC CURVES - A NONPARAMETRIC APPROACH [J].
DELONG, ER ;
DELONG, DM ;
CLARKEPEARSON, DI .
BIOMETRICS, 1988, 44 (03) :837-845
[3]   Partial AUC estimation and regression [J].
Dodd, LE ;
Pepe, MS .
BIOMETRICS, 2003, 59 (03) :614-623
[4]  
Emir B, 2000, STAT MED, V19, P511, DOI 10.1002/(SICI)1097-0258(20000229)19:4<511::AID-SIM353>3.0.CO
[5]  
2-3
[6]  
Gluer Claus-C, 2003, Curr Osteoporos Rep, V1, P98, DOI 10.1007/BF02686913
[7]   THE MEANING AND USE OF THE AREA UNDER A RECEIVER OPERATING CHARACTERISTIC (ROC) CURVE [J].
HANLEY, JA ;
MCNEIL, BJ .
RADIOLOGY, 1982, 143 (01) :29-36
[8]   A Procedure for Determining Whether a Simple Combination of Diagnostic Tests May Be Noninferior to the Theoretical Optimum Combination [J].
Jin, Hua ;
Lu, Ying .
MEDICAL DECISION MAKING, 2008, 28 (06) :909-916
[9]   Detection of hepatic metastases from cancers of the gastrointestinal tract by using noninvasive imaging methods (US, CT, MR imaging, PET): A meta-analysis [J].
Kinkel, K ;
Lu, Y ;
Both, M ;
Warren, RS ;
Thoeni, RF .
RADIOLOGY, 2002, 224 (03) :748-756
[10]   A non-inferiority test for diagnostic accuracy based on the paired partial areas under ROC curves [J].
Li, Chi-Rong ;
Liao, Chen-Tuo ;
Liu, Jen-Pei .
STATISTICS IN MEDICINE, 2008, 27 (10) :1762-1776
123