System of Linear Equations to Derive Unreported Test Accuracy Counts for Meta-Analysis.
Uloženo v:
| Název: | System of Linear Equations to Derive Unreported Test Accuracy Counts for Meta-Analysis. |
|---|---|
| Autoři: | Xie X; Acute and Hospital-Based Care, Ontario Health, Toronto, Ontario, Canada., Wang M; Acute and Hospital-Based Care, Ontario Health, Toronto, Ontario, Canada., Antony J; Acute and Hospital-Based Care, Ontario Health, Toronto, Ontario, Canada., Vandersluis S; Acute and Hospital-Based Care, Ontario Health, Toronto, Ontario, Canada., Kabali CB; Acute and Hospital-Based Care, Ontario Health, Toronto, Ontario, Canada.; Division of Epidemiology, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada. |
| Zdroj: | Statistics in medicine [Stat Med] 2025 Dec; Vol. 44 (28-30), pp. e70336. |
| Způsob vydávání: | Journal Article |
| Jazyk: | English |
| Informace o časopise: | Publisher: Wiley Country of Publication: England NLM ID: 8215016 Publication Model: Print Cited Medium: Internet ISSN: 1097-0258 (Electronic) Linking ISSN: 02776715 NLM ISO Abbreviation: Stat Med Subsets: MEDLINE |
| Imprint Name(s): | Original Publication: Chichester ; New York : Wiley, c1982- |
| Výrazy ze slovníku MeSH: | Meta-Analysis as Topic* , Diagnostic Tests, Routine*/standards, Humans ; Linear Models ; Sensitivity and Specificity ; False Positive Reactions ; False Negative Reactions ; Computer Simulation |
| Abstrakt: | Meta-analyses assessing test accuracy typically require extracting true positive (TP), false negative (FN), false positive (FP), and true negative (TN) counts from each study, commonly organized in a 2 × 2 table. However, many published test accuracy studies do not report all of these counts, which can limit the ability of a meta-analysis to fully capture the available evidence on the screening or diagnostic accuracy of a given test. Fortunately, test accuracy studies often report sufficient parameters, such as sensitivity and specificity, that enable the estimation of unreported counts. The relationships between these commonly reported parameters and the unreported cell counts may be expressed mathematically and organized into a system of four linear equations. The basic principles of solving such systems using matrix methods are introduced, accompanied by examples illustrating the development and solution of linear systems with unknown TP, FN, TN, and TN counts. Approaches for handling rounding errors of reported test accuracy parameters in publications are also demonstrated. Additionally, methods for obtaining a bound solution are explored in scenarios where the solution for missing test accuracy counts results in a system with three linear equations and four unknowns, leading to non-unique solutions. Simulation studies are conducted to assess the performance of these methods, and practical guidance for their implementation is provided. The Microsoft Excel spreadsheets and SAS and R code for the examples presented in this article are available in the Supporting Information. (© 2025 John Wiley & Sons Ltd.) |
| References: | J. B. Reitsma, A. S. Glas, A. W. Rutjes, R. J. Scholten, P. M. Bossuyt, and A. H. Zwinderman, “Bivariate Analysis of Sensitivity and Specificity Produces Informative Summary Measures in Diagnostic Reviews,” Journal of Clinical Epidemiology 58, no. 10 (2005): 982–990, https://doi.org/10.1016/j.jclinepi.2005.02.022. C. M. Rutter and C. A. Gatsonis, “A Hierarchical Regression Approach to Meta‐Analysis of Diagnostic Test Accuracy Evaluations,” Statistics in Medicine 20, no. 19 (2001): 2865–2884, https://doi.org/10.1002/sim.942. X. Xie, A. Sinclair, and N. Dendukuri, “Evaluating the Accuracy and Economic Value of a New Test in the Absence of a Perfect Reference Test,” Research Synthesis Methods 8, no. 3 (2017): 321–332, https://doi.org/10.1002/jrsm.1243. H. Chu, S. Chen, and T. A. Louis, “Random Effects Models in a Meta‐Analysis of the Accuracy of Two Diagnostic Tests Without a Gold Standard,” Journal of the American Statistical Association 104, no. 486 (2009): 512–523, https://doi.org/10.1198/jasa.2009.0017. Ontario Health, “Level 2 Polysomnography for the Diagnosis of Sleep Disorders: A Health Technology Assessment,” Ontario Health Technology Assessment Series 24, no. 7 (2024): 1–157. M. El Shayeb, L. A. Topfer, T. Stafinski, L. Pawluk, and D. Menon, “Diagnostic Accuracy of Level 3 Portable Sleep Tests Versus Level 1 Polysomnography for Sleep‐Disordered Breathing: A Systematic Review and Meta‐Analysis,” CMAJ 186, no. 1 (2014): E25–E51, https://doi.org/10.1503/cmaj.130952. M. Spick, H. M. Lewis, M. J. Wilde, C. Hopley, J. Huggett, and M. J. Bailey, “Systematic Review With Meta‐Analysis of Diagnostic Test Accuracy for COVID‐19 by Mass Spectrometry,” Metabolism 126 (2022): 154922, https://doi.org/10.1016/j.metabol.2021.154922. SAS Institute Inc, “Base SAS 9.4,” (2020), Cary, NC SAS Institute Inc. “R: A Language and Environment for Statistical Computing [Computer Program],” (2023), Vienna: R Foundation for Statistical Computing. H. Anton and C. Rorres, Elementary Linear Algebra: Applications Version, 7th ed. (Wiley, 1994). T. Asano, M. Takemura, K. Fukumitsu, et al., “Diagnostic Utility of Fractional Exhaled Nitric Oxide in Prolonged and Chronic Cough According to Atopic Status,” Allergology International 66, no. 2 (2017): 344–350, https://doi.org/10.1016/j.alit.2016.08.015. T. P. Morris, I. R. White, and M. J. Crowther, “Using Simulation Studies to Evaluate Statistical Methods,” Statistics in Medicine 38, no. 11 (2019): 2074–2102, https://doi.org/10.1002/sim.8086. H. Zeisler, E. Llurba, F. Chantraine, et al., “Predictive Value of the sFlt‐1:PlGF Ratio in Women With Suspected Preeclampsia,” New England Journal of Medicine 374, no. 1 (2016): 13–22, https://doi.org/10.1056/NEJMoa1414838. F. Alarid‐Escudero, R. F. MacLehose, Y. Peralta, K. M. Kuntz, and E. A. Enns, “Nonidentifiability in Model Calibration and Implications for Medical Decision Making,” Medical Decision Making 38, no. 7 (2018): 810–821, https://doi.org/10.1177/0272989X18792283. J. M. Porcel, A. Ruiz‐Gonzalez, M. Falguera, et al., “Contribution of a Pleural Antigen Assay (Binax NOW) to the Diagnosis of Pneumococcal Pneumonia,” Chest 131, no. 5 (2007): 1442–1447, https://doi.org/10.1378/chest.06‐1884. J. P. Vielma, “Mixed Integer Linear Programming Formulation Techniques,” SIAM Review 57, no. 1 (2015): 3–57. H. E. Jones, C. A. Gatsonsis, T. A. Trikalinos, N. J. Welton, and A. E. Ades, “Quantifying How Diagnostic Test Accuracy Depends on Threshold in a Meta‐Analysis,” Statistics in Medicine 38, no. 24 (2019): 4789–4803, https://doi.org/10.1002/sim.8301. A. Zapf, C. Albert, C. Fromke, et al., “Meta‐Analysis of Diagnostic Accuracy Studies With Multiple Thresholds: Comparison of Different Approaches,” Biometrical Journal 63, no. 4 (2021): 699–711, https://doi.org/10.1002/bimj.202000091. R. D. Riley, I. Ahmed, J. Ensor, et al., “Meta‐Analysis of Test Accuracy Studies: An Exploratory Method for Investigating the Impact of Missing Thresholds,” Systematic Reviews 4 (2015): 12, https://doi.org/10.1186/2046‐4053‐4‐12. S. Steinhauser, M. Schumacher, and G. Rucker, “Modelling Multiple Thresholds in Meta‐Analysis of Diagnostic Test Accuracy Studies,” BMC Medical Research Methodology 16, no. 1 (2016): 97, https://doi.org/10.1186/s12874‐016‐0196‐1. |
| Contributed Indexing: | Keywords: 2 × 2 table; meta‐analysis; system of linear equations; test accuracy |
| Entry Date(s): | Date Created: 20251203 Date Completed: 20251203 Latest Revision: 20251203 |
| Update Code: | 20251203 |
| DOI: | 10.1002/sim.70336 |
| PMID: | 41332421 |
| Databáze: | MEDLINE |
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstrakt: | Meta-analyses assessing test accuracy typically require extracting true positive (TP), false negative (FN), false positive (FP), and true negative (TN) counts from each study, commonly organized in a 2 × 2 table. However, many published test accuracy studies do not report all of these counts, which can limit the ability of a meta-analysis to fully capture the available evidence on the screening or diagnostic accuracy of a given test. Fortunately, test accuracy studies often report sufficient parameters, such as sensitivity and specificity, that enable the estimation of unreported counts. The relationships between these commonly reported parameters and the unreported cell counts may be expressed mathematically and organized into a system of four linear equations. The basic principles of solving such systems using matrix methods are introduced, accompanied by examples illustrating the development and solution of linear systems with unknown TP, FN, TN, and TN counts. Approaches for handling rounding errors of reported test accuracy parameters in publications are also demonstrated. Additionally, methods for obtaining a bound solution are explored in scenarios where the solution for missing test accuracy counts results in a system with three linear equations and four unknowns, leading to non-unique solutions. Simulation studies are conducted to assess the performance of these methods, and practical guidance for their implementation is provided. The Microsoft Excel spreadsheets and SAS and R code for the examples presented in this article are available in the Supporting Information.<br /> (© 2025 John Wiley & Sons Ltd.) |
|---|---|
| ISSN: | 1097-0258 |
| DOI: | 10.1002/sim.70336 |