Estimators of sensitivity and specificity in the presence of verification bias: A Bayesian approach

Verification bias can occur if some of the patients with test results are not selected to receive the gold standard procedure. Unverified cases frequently are not suggestive to be positives. Consequently, the set of verified cases overestimates the number of true positives and underestimates the number of true negatives. The sensitivity and specificity estimates based only on the patients with verified disease are often biased. In this article we derive estimators for sensitivity and specificity not subject to verification bias using a Bayesian approach. Marginal posterior densities of all parameters are estimated using the Gibbs sampler algorithm. An application to the study of accuracy of Hybrid Capture II in the diagnosis of cervical intraepithelial neoplasia grades 2 and 3 illustrates the proposed methodology.

[1]  D. Heitjan,et al.  Distinguishing “Missing at Random” and “Missing Completely at Random” , 1996 .

[2]  Xiao-Hua Zhou,et al.  Nonparametric analysis for the ROC areas of two diagnostic tests in the presence of nonignorable verification bias , 2003 .

[3]  P. Bossuyt,et al.  Sources of Variation and Bias in Studies of Diagnostic Accuracy , 2004, Annals of Internal Medicine.

[4]  D. Altman,et al.  Statistics Notes: Diagnostic tests 2: predictive values , 1994, BMJ.

[5]  Bradley P. Carlin,et al.  BAYES AND EMPIRICAL BAYES METHODS FOR DATA ANALYSIS , 1996, Stat. Comput..

[6]  Huiman X Barnhart,et al.  Evaluating medical diagnostic tests at the subunit level in the presence of verification bias. , 2003, Statistics in medicine.

[7]  Todd A Alonzo,et al.  Verification bias‐corrected estimators of the relative true and false positive rates of two binary screening tests , 2005, Statistics in medicine.

[8]  C B Begg,et al.  Biases in the assessment of diagnostic tests. , 1987, Statistics in medicine.

[9]  C. Roehrborn,et al.  Verification and incorporation biases in studies assessing screening tests: prostate-specific antigen as an example. , 2004, Urology.

[10]  Bradley P. Carlin,et al.  BAYES AND EMPIRICAL BAYES METHODS FOR DATA ANALYSIS , 1996, Stat. Comput..

[11]  D. Altman,et al.  Diagnostic tests 4: likelihood ratios , 2004, BMJ : British Medical Journal.

[12]  R. Little.  A Test of Missing Completely at Random for Multivariate Data with Missing Values , 1988 .

[13]  Geert Molenberghs,et al.  Likelihood Based Frequentist Inference When Data Are Missing at Random , 1998 .

[14]  Xiao-Hua Zhou,et al.  Maximum likelihood estimators of sensitivity and specificity corrected for verification bias , 1993 .

[15]  X H Zhou,et al.  Assessing the relative accuracies of two screening tests in the presence of verification bias. , 2000, Statistics in medicine.

[16]  Eugene H Blackstone,et al.  Caveat emptor: the treachery of work-up bias. , 2004, The Journal of thoracic and cardiovascular surgery.

[17]  R A Greenes,et al.  Assessment of diagnostic tests when disease verification is subject to selection bias. , 1983, Biometrics.

[18]  T. Louis,et al.  Bayes and Empirical Bayes Methods for Data Analysis. , 1997 .

[19]  Adrian F. M. Smith,et al.  Sampling-Based Approaches to Calculating Marginal Densities , 1990 .

[20]  Xiao-Hua Zhou,et al.  Estimating an roc curve in the presence of non-ignorable verification bias , 1998 .

[21]  X H Zhou,et al.  Correcting for verification bias in studies of a diagnostic test's accuracy , 1998, Statistical methods in medical research.

[22]  G. C. Tiao,et al.  Bayesian inference in statistical analysis , 1973 .

[23]  D. Altman,et al.  Statistics Notes: Diagnostic tests 1: sensitivity and specificity , 1994 .

[24]  P M Bossuyt,et al.  Effect of study design on the association between nuchal translucency measurement and Down syndrome. , 1999, Obstetrics and gynecology.

[25]  Huiman X Barnhart,et al.  Accounting for Nonignorable Verification Bias in Assessment of Diagnostic Tests , 2003, Biometrics.

[26]  Xiao-Hua Zhou,et al.  Adjusting for non‐ignorable verification bias in clinical studies for Alzheimer's disease , 2004, Statistics in medicine.

[27]  B. McNeil,et al.  Assessment of radiologic tests: control of bias and other design considerations. , 1988, Radiology.

[28]  W. Wong,et al.  The calculation of posterior distributions by data augmentation , 1987 .

[29]  Edson Zangiacomi Martinez,et al.  Desempenho do exame colpocitológico com revisão por diferentes observadores e da captura híbrida II no diagnóstico da neoplasia intra-epitelial cervical graus 2 e 3 , 2003 .