Quantifying how diagnostic test accuracy depends on threshold in a meta‐analysis
Open Access
- 30 September 2019
- journal article
- research article
- Published by Wiley in Statistics in Medicine
- Vol. 38 (24) , 4789-4803
Abstract
Tests for disease often produce a continuous measure, such as the concentration of some biomarker in a blood sample. In clinical practice, a thresholdCis selected such that results, say, greater thanCare declared positive and those less thanCnegative. Measures of test accuracy such as sensitivity and specificity depend crucially onC, and the optimal value of this threshold is usually a key question for clinical practice. Standard methods for meta‐analysis of test accuracy (i) do not provide summary estimates of accuracy at each threshold, precluding selection of the optimal threshold, and furthermore, (ii) do not make use of all available data. We describe a multinomial meta‐analysis model that can take any number of pairs of sensitivity and specificity from each study and explicitly quantifies how accuracy depends onC. Our model assumes that some prespecified or Box‐Cox transformation of test results in the diseased and disease‐free populations has a logistic distribution. The Box‐Cox transformation parameter can be estimated from the data, allowing for a flexible range of underlying distributions. We parameterise in terms of the means and scale parameters of the two logistic distributions. In addition to credible intervals for the pooled sensitivity and specificity across all thresholds, we produce prediction intervals, allowing for between‐study heterogeneity in all parameters. We demonstrate the model using two case study meta‐analyses, examining the accuracy of tests for acute heart failure and preeclampsia. We show how the model can be extended to explore reasons for heterogeneity using study‐level covariates.Keywords
Funding Information
- Medical Research Council (MR/M014533/1)
- National Institute for Health Research
This publication has 38 references indexed in Scilit:
- Multivariate random effects meta-analysis of diagnostic tests with multiple thresholdsBMC Medical Research Methodology, 2009
- A Re-Evaluation of Random-Effects Meta-AnalysisJournal of the Royal Statistical Society Series A: Statistics in Society, 2008
- Diagnostic and Prognostic Utility of Brain Natriuretic Peptide in Subjects Admitted to the ICU With Hypoxic Respiratory Failure Due to Noncardiogenic and Cardiogenic Pulmonary EdemaChest, 2007
- Influence of history of heart failure on diagnostic performance and utility of B-type natriuretic peptide testing for acute dyspnea in the emergency departmentAmerican Heart Journal, 2006
- The Inconsistency of “Optimal” Cutpoints Obtained using Two Criteria based on the Receiver Operating Characteristic CurveAmerican Journal of Epidemiology, 2006
- Elecsys NT-ProBNP and BNP Assays: Are There Analytically and Clinically Relevant Differences?Journal of Cardiac Failure, 2005
- Comparative accuracy of B-type natriuretic peptide and tissue Doppler echocardiography in the diagnosis of congestive heart failureThe American Journal of Cardiology, 2004
- Bayesian Measures of Model Complexity and FitJournal of the Royal Statistical Society Series B: Statistical Methodology, 2002
- How should meta‐regression analyses be undertaken and interpreted?Statistics in Medicine, 2002
- Two transformation models for estimating an ROC curve derived from continuous dataJournal of Applied Statistics, 2000