An extended trivariate vine copula mixed model for meta-analysis of diagnostic studies in the presence of non-evaluable outcomes
-
Aristidis K. Nikoloulopoulos
Abstract
A recent paper proposed an extended trivariate generalized linear mixed model (TGLMM) for synthesis of diagnostic test accuracy studies in the presence of non-evaluable index test results. Inspired by the aforementioned model we propose an extended trivariate vine copula mixed model that includes the TGLMM as special case, but can also operate on the original scale of sensitivity, specificity, and disease prevalence. The performance of the proposed vine copula mixed model is examined by extensive simulation studies in comparison with the TGLMM. Simulation studies showed that the TGLMM leads to biased meta-analytic estimates of sensitivity, specificity, and prevalence when the univariate random effects are misspecified. The vine copula mixed model gives nearly unbiased estimates of test accuracy indices and disease prevalence. Our general methodology is illustrated by meta-analysing coronary CT angiography studies.
Funding source: University of East Anglia
Acknowledgements
The simulations presented in this paper were carried out on the High Performance Computing Cluster supported by the Research and Specialist Computing Support service at the University of East Anglia.
-
Author contribution: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
-
Research funding: None declared.
-
Conflict of interest statement: The authors declare no conflicts of interest regarding this article.
References
1. Jackson, D, Riley, R, White, IR. Multivariate meta-analysis: Potential and promise. Stat Med 2011;30:2481–98.10.1002/sim.4172Search in Google Scholar
2. Mavridis, D, Salanti, G. A practical introduction to multivariate meta-analysis. Stat Method Med Res 2013;22:133–58.10.1177/0962280211432219Search in Google Scholar
3. Ma, X, Nie, L, R Cole, S, Chu, H. Statistical methods for multivariate meta-analysis of diagnostic tests: An overview and tutorial. Stat Method Med Res 2016;25:1596–619.10.1177/0962280213492588Search in Google Scholar
4. Begg, CB, Greenes, RA, Iglewicz, B. The influence of uninterpretability on the assessment of diagnostic tests. J Chron Dis 1986;39:575–84.10.1016/0021-9681(86)90182-7Search in Google Scholar
5. Schuetz, GM, Schlattmann, P, Dewey, M. Use of 3×2 tables with an intention to diagnose approach to assess clinical performance of diagnostic tests: Meta-analytical evaluation of coronary CT angiography studies. BMJ (Online) 2012;345:e6717.10.1136/bmj.e6717Search in Google Scholar
6. Ma, X, Suri, MFK, Chu, H. A trivariate meta-analysis of diagnostic studies accounting for prevalence and non-evaluable subjects: Re-evaluation of the meta-analysis of coronary CT angiography studies. BMC Med Res Methodol 2014;14:128.10.1186/1471-2288-14-128Search in Google Scholar
7. Chu, H, Cole, SR. Bivariate meta-analysis of sensitivity and specificity with sparse data: A generalized linear mixed model approach. J Clin Epidemiol 2006;59:1331–2.10.1016/j.jclinepi.2006.06.011Search in Google Scholar
8. Chu, H, Nie, L, Cole, SR, Poole, C. Meta-analysis of diagnostic accuracy studies accounting for disease prevalence: Alternative parameterizations and model selection. Stat Med 2009;28:2384–99.10.1002/sim.3627Search in Google Scholar
9. Brenner, H, Gefeller, O. Variation of sensitivity, specificity, likelihood ratios and predictive values with disease prevalence. Stat Med 1997;16:981–91.10.1002/(SICI)1097-0258(19970515)16:9<981::AID-SIM510>3.0.CO;2-NSearch in Google Scholar
10. Leeflang, MMG, Bossuyt, PMM, Irwig, L. Diagnostic test accuracy may vary with prevalence: Implications for evidence-based diagnosis. J Clin Epidemiol 2009;62:5–12.10.1016/j.jclinepi.2008.04.007Search in Google Scholar
11. Leeflang, MMG, Rutjes, AWS, Reitsma, JB, Hooft, L, Bossuyt, PMM. Variation of a test's sensitivity and specificity with disease prevalence. Canad Med Assoc J 2013;185:E537–44.10.1503/cmaj.121286Search in Google Scholar PubMed PubMed Central
12. Nikoloulopoulos, AK. A vine copula mixed effect model for trivariate meta-analysis of diagnostic test accuracy studies accounting for disease prevalence. Stat Method Med Res 2017;26:2270–86.10.1177/0962280215596769Search in Google Scholar PubMed
13. Joe, H, Li, H, Nikoloulopoulos, AK. Tail dependence functions and vine copulas. J Multivariate Anal 2010;101:252–70.10.1016/j.jmva.2009.08.002Search in Google Scholar
14. Aas, K, Czado, C, Frigessi, A, Bakken, H. Pair-copula constructions of multiple dependence. Ins Math Econ 2009;44:182–98.10.1016/j.insmatheco.2007.02.001Search in Google Scholar
15. Joe, H. Families of m-variate distributions with given margins and m(m-1)/2 bivariate dependence parameters. In Rüschendorf, L, Schweizer, B, Taylor, MD, editors, Distributions with Fixed Marginals and Related Topics, vol 28, 120–41, Hayward, CA: Institute of Mathematical Statistics; 1996.10.1214/lnms/1215452614Search in Google Scholar
16. Nikoloulopoulos, AK. A mixed effect model for bivariate meta-analysis of diagnostic test accuracy studies using a copula representation of the random effects distribution. Stat Med 2015;34:3842–65.10.1002/sim.6595Search in Google Scholar PubMed
17. Nikoloulopoulos, AK. Hybrid copula mixed models for combining case-control and cohort studies in meta-analysis of diagnostic tests. Stat Method Med Res 2018;27:2540–53.10.1177/0962280216682376Search in Google Scholar PubMed
18. Nikoloulopoulos, AK. On composite likelihood in bivariate meta-analysis of diagnostic test accuracy studies. AStA Adv Stat Anal 2018;102:211–27.10.1007/s10182-017-0299-ySearch in Google Scholar
19. Nikoloulopoulos, AK. A D-vine copula mixed model for joint meta-analysis and comparison of diagnostic tests. Stat Method Med Res 2019;28:3286–300.10.1177/0962280218796685Search in Google Scholar PubMed
20. Nash, JC. Compact Numerical Methods for Computers: Linear Algebra and Function Minimisation, 2nd edition. New York: Hilger; 1990.Search in Google Scholar
21. Stroud, AH, Secrest, D. Gaussian Quadrature Formulas. Englewood Cliffs, NJ: Prentice-Hall; 1966.Search in Google Scholar
22. Little, RJ, Rubin, D. Statistical Analysis with Missing Data. Hoboken, NJ: John Wiley & Sons; 2002.10.1002/9781119013563Search in Google Scholar
23. Joe, H. Dependence comparisons of vine copulae with four or more variables. In Kurowicka, D, Joe, H, editors, Dependence Modeling: Handbook on Vine Copulae, chapter 7, 139–64. Singapore: World Scientific; 2011.10.1142/9789814299886_0007Search in Google Scholar
24. Hult, H, Lindskog, F. Multivariate extremes, aggregation and dependence in elliptical distributions. Adv Appl Prob 2002;34:587–608.10.1239/aap/1033662167Search in Google Scholar
25. Genest, C. Frank's family of bivariate distributions. Biometrika 1987;74:549–55.10.1093/biomet/74.3.549Search in Google Scholar
26. Genest, C, MacKay, J. The joy of copulas: Bivariate distributions with uniform marginals. Am Stat 1986;40:280–3.10.1080/00031305.1986.10475414Search in Google Scholar
27. Paul, M, Riebler, A, Bachmann, LM, Rue, H, Held, L. Bayesian bivariate meta-analysis of diagnostic test studies using integrated nested laplace approximations. Stat Med 2010;29:1325–39.10.1002/sim.3858Search in Google Scholar PubMed
28. Nikoloulopoulos, AK. A multinomial quadrivariate D-vine copula mixed model for meta-analysis of diagnostic studies in the presence of non-evaluable subjects. Stat Methods Med Res 2020. https://doi.org/10.1177/0962280220913898.10.1177/0962280220913898Search in Google Scholar PubMed PubMed Central
29. Nikoloulopoulos, AK. CopulaREMADA: copula mixed models for multivariate meta-analysis of diagnostic test accuracy studies, 2019. R package version 1.3. Available from: http://CRAN.R-project.org/package=CopulaREMADA.Search in Google Scholar
Supplementary Material
The online version of this article offers supplementary material (https://doi.org/10.1515/ijb-2019-0107).
© 2020 Walter de Gruyter GmbH, Berlin/Boston
Articles in the same Issue
- Research Articles
- Nonparametric bootstrap inference for the targeted highly adaptive least absolute shrinkage and selection operator (LASSO) estimator
- A Bayesian Framework for Robust Quantitative Trait Locus Mapping and Outlier Detection
- Inference for the Analysis of Ordinal Data with Spatio-Temporal Models
- An iterative algorithm for joint covariate and random effect selection in mixed effects models
- An extended trivariate vine copula mixed model for meta-analysis of diagnostic studies in the presence of non-evaluable outcomes
- Super Learner for Survival Data Prediction
- Model-based random forests for ordinal regression
- A Parametric Bootstrap for the Mean Measure of Divergence
- Derivation of Passing–Bablok regression from Kendall’s tau
- Direct effect and indirect effect on an outcome under nonlinear modeling
- GAMLSS for high-variability data: an application to liver fibrosis case
- Variable selection for high-dimensional quadratic Cox model with application to Alzheimer’s disease
Articles in the same Issue
- Research Articles
- Nonparametric bootstrap inference for the targeted highly adaptive least absolute shrinkage and selection operator (LASSO) estimator
- A Bayesian Framework for Robust Quantitative Trait Locus Mapping and Outlier Detection
- Inference for the Analysis of Ordinal Data with Spatio-Temporal Models
- An iterative algorithm for joint covariate and random effect selection in mixed effects models
- An extended trivariate vine copula mixed model for meta-analysis of diagnostic studies in the presence of non-evaluable outcomes
- Super Learner for Survival Data Prediction
- Model-based random forests for ordinal regression
- A Parametric Bootstrap for the Mean Measure of Divergence
- Derivation of Passing–Bablok regression from Kendall’s tau
- Direct effect and indirect effect on an outcome under nonlinear modeling
- GAMLSS for high-variability data: an application to liver fibrosis case
- Variable selection for high-dimensional quadratic Cox model with application to Alzheimer’s disease
