The accuracy of a diagnostic test with continuous-scale results is of high importance in clinical medicine. It is often summarised by the area under the ROC curve (AUC). In this article, we discuss and compare nine non-parametric confidence intervals of the AUC for a continuous-scale diagnostic test. Simulation studies are conducted to evaluate the relative performance of the confidence intervals for the AUC in terms of coverage probability and average interval length. A real example is used to illustrate the application of the recommended methods.
Get full access to this article
View all access options for this article.
References
1.
Zhou XH, Obuchowski NA McClish DMStatistical Methods in Diagnostic Medicine. Wiley & Sons, 2002.
2.
Pepe MSThe Statistical Evaluation of Medical Tests for Classification and Prediction . Oxford University Press, 2003.
3.
Hanley JAReceiver operating characteristic (ROC) methodology: the state of the art . Clinical Reviews in Diagnostic Imaging1989 ; 29: 307-35.
4.
Swets JAThe relative operating characteristic in psychology. Science1973; 182: 990-1000.
5.
Swets JAMeasuring the accuracy of diagnostic systems. Science1988; 240: 1285-93.
6.
Shapiro DEThe interpretation of diagnostic tests. Statistical Methods in Medical Research1999; 8: 113-34.
7.
Bamber DCThe area above the ordinal dominance graph and the area below the receiver operating characteristic curve graph. J Math Psychology1975; 12: 387-415.
8.
Wolfe DA, Hogg RVOn constructing statistics and reporting data. American Statistician1971; 25: 27-30.
9.
Wieand S., Gail MH, James BR, James KRA family of non-parametric statistics for comparing diagnostic markers with paired and unpaired data. Biometrika1989; 76: 585-92.
10.
Metz CE, Herman BA, Shen J.-H.Maximum likelihood estimation of receiver operating characteristic (ROC) curves from continuously-distributed data. Statistics in Medicine1998; 17: 1033-53.
11.
Tong H.On the estimation of P(Y < X) for exponential families. IEEE Transactions on Reliability1977; 26: 54-56.
12.
Pham T., Almhana J.The generalized gamma distribution: its hazard rate and stress-strength model . IEEE Transactions on Reliability1995 ; 44: 392-97.
13.
Gupta CG, Brown N.Reliability studies of the skew-normal distribution and its application to a stress-strength model. Communications in Statistics: Theory and Methods2001; 30: 2427-45.
14.
Obuchowski NA, Lieber MLConfidence bounds when estimated ROC area is 1.0. Academic Radiology2002; 9: 526-30.
15.
Obuchowski NA, Lieber MLConfidence intervals for the receiver operating characteristic area in studies with small samples. Academic Radiology1998; 5: 561-71.
16.
Hanley JA, Hajian-Tilaki KOSampling variability of non-parametric estimates of the area under receiver operating characteristic curves: an update. Academic Radiology1997; 4: 49-58.
17.
DeLong EA , DeLong DM, Clarke-Pearson DLComparing the area under receiver operating characteristic curves: a nonparametric approach. Biometrics1988; 44: 837-45.
18.
Sen PKA note on asymptotically distribution-free confidence bounds for P(X < Y) based on two independent samples. Sankhyã, Series A1967; 29: 95-102.
19.
Qin GS, Zhou XHEmpirical likelihood inference for the area under the ROC curve. Biometrics2006; 62: 613-22.
20.
Rao ACR , Collinson PO, Canepa-Anson R., Josepha SPTroponin T measurement after myocardial infarction can identify left ventricular ejection of less than 40%. Heart1998; 80: 223-25.
21.
Pepe MS, Cai T.The analysis of placement values for evaluating discriminatory measures . Biometrics2004; 60: 528-35.
22.
Efron B., Tibshirani RJAn Introduction to the Bootstrap. Chapman and Hall , 1993.
23.
Shao J., Tu D.The Jackknife and Bootstrap. Springer-Verlag, 1995.
24.
Zhou XH, Qin GSImproved confidence intervals for the sensitivity of a receiver operating characteristic-scale test at a fixed level of specificity. Statistics in Medicine2005; 24: 465-77. DOI:10.1002/sim.1563
