Estimation of diagnostic test accuracy and disease prevalence in the absence of a perfect reference test
A perfect diagnostic test (with 100% sensitivity and specificity) is for the most part a theoretical concept. In practice, there are several diseases (e.g. tuberculosis, pneumonia, Alzheimer's disease) for which there is no perfect test that can detect the presence of the disease with certainty. This complicates estimation of disease prevalence as well as the evaluation of new diagnostic tests. Latent class models have been proposed as a possible solution for this problem as they adjust for the imperfect accuracy of all tests involved.
Some of my work in this area is related to the extension of the traditional latent class model in several directions, including modeling conditional dependence between multiple diagnostic tests and adjustment for verification bias. Bayesian estimation is very useful for these models which can sometimes be non-identifiable and are typically quite complex.
Other work pertains to applying Bayesian sample size methods for designing studies using non-gold standard diagnostic tests. These papers have illustrated that these types of studies need sample sizes that are much larger than those typically encountered in diagnostic studies.
Finally, some publications have been related to illustrating that simpler alternatives to latent class analysis, such as the composite reference standards, are problematic.
Latent class models
- de Groot JA, Dendukuri N, Janssen KJ, Reitsma JB, Brophy J, Joseph L, Bossuyt PM, Moons KG. Adjusting for partial verification or workup bias in meta-analyses of diagnostic accuracy studies. Am J Epidemiol. 2012 Apr 15;175(8):847-53.
- Lu Y, Dendukuri N, Schiller I, Joseph L. A Bayesian approach to simultaneously adjusting for verification and reference standard bias in diagnostic test studies. Stat Med. 2010 Oct 30;29(24):2532-43.
- Ling DI, Pai M, Schiller I, Dendukuri N. A Bayesian framework for estimating the incremental value of a diagnostic test in the absence of a gold standard. BMC Med Res Methodol. 2014 May 15;14:67.
Sample size estimation
- Dendukuri N, Rahme E, Bélisle P, Joseph L. Bayesian sample size determination for prevalence and diagnostic test studies in the absence of a gold standard test. Biometrics. 2004 Jun;60(2):388-97.
- Hadgu A, Dendukuri N, Hilden J. Evaluation of nucleic acid amplification tests in the absence of a perfect gold-standard test: a review of the statistical and epidemiologic issues. Epidemiology. 2005 Sep;16(5):604-12.
- Pai M, Dendukuri N, Wang L, Joshi R, Kalantri S, Rieder HL. Improving the estimation of tuberculosis infection prevalence using T-cell-based assay and mixture models. Int J Tuberc Lung Dis. 2008 Aug;12(8):895-902.
- Buczinski S, L Ollivett T, Dendukuri N. Bayesian estimation of the accuracy of the calf respiratory scoring chart and ultrasonography for the diagnosis of bovine respiratory disease in pre-weaned dairy calves. Prev Vet Med. 2015 May1;119(3-4):227-31.
Problems with composite reference standards
- Schiller I, van Smeden M, Hadgu A, Libman M, Reitsma JB, Dendukuri N. Bias due to composite reference standards in diagnostic accuracy studies. Statistics in Medicine. 2015. (link)
Applications of latent class models
- Dendukuri N, Joseph L. Bayesian approaches to modeling the conditional dependence between multiple diagnostic tests. Biometrics. 2001 Mar;57(1):158-67.
- Dendukuri N, Hadgu A, Wang L. Modeling conditional dependence between diagnostic tests: a multiple latent variable model. Stat Med. 2009 Feb 1;28(3):441-61.
- Dendukuri N, Schiller I, Joseph L, Pai M. Bayesian meta-analysis of the accuracy of a test for tuberculous pleuritis in the absence of a gold standard reference. Biometrics. 2012 Dec;68(4):1285-93.
- Dendukuri N, Bélisle P, Joseph L. Bayesian sample size for diagnostic test studies in the absence of a gold standard: Comparing identifiable with non-identifiable models. Stat Med. 2010 Nov 20;29(26):2688-97.
- Hadgu A, Dendukuri N, Wang L. Evaluation of screening tests for detecting Chlamydia trachomatis: bias associated with the patient-infected-status algorithm. Epidemiology. 2012