How to deal with double partial verification when evaluating two index tests in relation to a reference test?

Research into the diagnostic accuracy of clinical tests is often hampered by single or double partial verification mechanisms, that is, not all patients have their disease status verified by a reference test, neither do all patients receive all tests under evaluation (index tests). We show methods that reduce verification bias introduced when omitting data from partially tested patients. Adjustment techniques are well established when there are no missing index tests and when the reference test is 'missing at random'. However, in practice, index tests tend to be omitted, and the choice of applying a reference test may depend on unobserved variables related to disease status, that is, verification may be missing not at random (MNAR). We study double partial verification in a clinical example from reproductive medicine in which we analyse the diagnostic values of the chlamydia antibody test and the hysterosalpingography in relation to a diagnostic laparoscopy. First, we plot all possible combinations of sensitivity and specificity of both index tests in two test ignorance regions. Then, we construct models in which we impose different assumptions for the verification process. We allow for missing index tests, study the influence of patient characteristics and study the accuracy estimates if an MNAR mechanism would operate. It is shown that data on tests used in the diagnostic process of the same population are preferably studied jointly and that the influence of an MNAR verification process was limited in a clinical study where more than half of the patients did not have the reference test.

[1]  S G Baker,et al.  Evaluating multiple diagnostic tests with partial verification. , 1995, Biometrics.

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

[3]  Huiman X Barnhart,et al.  A global sensitivity analysis of performance of a medical diagnostic test when verification bias is present , 2003, Statistics in medicine.

[4]  B. Mol,et al.  Are patient characteristics associated with the accuracy of hysterosalpingography in diagnosing tubal pathology? An individual patient data meta-analysis. , 2011, Human reproduction update.

[5]  Johannes B. Reitsma,et al.  A review of solutions for diagnostic accuracy studies with an imperfect or missing reference standard. , 2009, Journal of clinical epidemiology.

[6]  Xiao-Hua Zhou,et al.  Comparing accuracies of two screening tests in a two‐phase study for dementia , 2002 .

[7]  Dankmar Böhning,et al.  A Capture–Recapture Approach for Screening Using Two Diagnostic Tests With Availability of Disease Status for the Test Positives Only , 2008 .

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

[9]  J A Land,et al.  Chlamydia antibody testing and diagnosing tubal pathology in subfertile women: an individual patient data meta-analysis. , 2011, Human reproduction update.

[10]  James M. Robins,et al.  Semiparametric Regression for Repeated Outcomes With Nonignorable Nonresponse , 1998 .

[11]  K J M Janssen,et al.  Multiple imputation to correct for partial verification bias revisited , 2008, Statistics in medicine.

[12]  P. Bossuyt,et al.  Identifying subfertile ovulatory women for timely tubal patency testing: a clinical decision rule based on medical history. , 2007, Human reproduction.

[13]  L. Dodd,et al.  On Estimating Diagnostic Accuracy From Studies With Multiple Raters and Partial Gold Standard Evaluation , 2008, Journal of the American Statistical Association.

[14]  D Scharfstein,et al.  Methods for Conducting Sensitivity Analysis of Trials with Potentially Nonignorable Competing Causes of Censoring , 2001, Biometrics.

[15]  Comparison of the accuracy of multiple binary tests in the presence of partial disease verification , 2010 .

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

[17]  P. Albert,et al.  A Cautionary Note on the Robustness of Latent Class Models for Estimating Diagnostic Error without a Gold Standard , 2004, Biometrics.

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

[19]  Geert Molenberghs,et al.  Sensitivity analysis for incomplete contingency tables: the Slovenian plebiscite case , 2001 .

[20]  Paul S Albert,et al.  Estimating diagnostic accuracy of multiple binary tests with an imperfect reference standard , 2009, Statistics in medicine.

[21]  M. Kenward,et al.  Every missingness not at random model has a missingness at random counterpart with equal fit , 2008 .

[22]  J W Hogan,et al.  Reparameterizing the Pattern Mixture Model for Sensitivity Analyses Under Informative Dropout , 2000, Biometrics.

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

[24]  A. Zwinderman,et al.  Correcting for partial verification bias: a comparison of methods. , 2011, Annals of epidemiology.

[25]  Joseph G. Ibrahim,et al.  Missing covariates in generalized linear models when the missing data mechanism is non‐ignorable , 1999 .

[26]  A. Agresti,et al.  Approximate is Better than “Exact” for Interval Estimation of Binomial Proportions , 1998 .

[27]  Geert Molenberghs,et al.  EVERY MISSING NOT AT RANDOM MODEL HAS GOT A MISSING AT RANDOM COUNTERPART WITH EQUAL FIT , 2008 .

[28]  Xiao-Hua Zhou,et al.  A model for adjusting for nonignorable verification bias in estimation of the ROC curve and its area with likelihood-based approach. , 2010, Biometrics.

[29]  A. Feinstein,et al.  Problems of spectrum and bias in evaluating the efficacy of diagnostic tests. , 1978, The New England journal of medicine.