When are summary ROC curves appropriate for diagnostic meta‐analyses?
Authors:
- 9 July 2009
- journal article
- research article
- Published by Wiley in Statistics in Medicine
- Vol. 28 (21) , 2653-2668
Abstract
Diagnostic tests are increasingly evaluated with systematic reviews and this has lead to the recent developments of statistical methods to analyse such data. The most commonly used method is the summary receiver operating characteristic (SROC) curve, which can be fitted with a non‐linear bivariate random‐effects model. This paper focuses on the practical problems of interpreting and presenting data from such analyses. First, many meta‐analyses may be underpowered to obtain reliable estimates of the SROC parameters. Second, the SROC model may be inappropriate. In these situations, a summary with two univariate meta‐analyses of the true and false positive rates (TPRs and FPRs) may be more appropriate. We characterize the type of problems that can occur in fitting these models and present an algorithm to guide the analyst of such studies, with illustrations from analyses of published data. A set of R functions, freely available to perform these analyses, can be downloaded from (www.diagmeta.info). Copyright © 2009 John Wiley & Sons, Ltd.Keywords
Funding Information
- NHS (CSO ref: CZF/1/23)
- Scottish Funding Council SINAPSE Initiative
This publication has 21 references indexed in Scilit:
- Bivariate Random Effects Meta-Analysis of ROC CurvesMedical Decision Making, 2008
- Flexible parametric models for random‐effects distributionsStatistics in Medicine, 2007
- Impact of Adjustment for Quality on Results of Metaanalyses of Diagnostic AccuracyClinical Chemistry, 2007
- Bivariate meta-analysis of sensitivity and specificity with sparse data: a generalized linear mixed model approachJournal of Clinical Epidemiology, 2006
- Bivariate analysis of sensitivity and specificity produces informative summary measures in diagnostic reviewsJournal of Clinical Epidemiology, 2005
- Empirical Bayes estimates generated in a hierarchical summary ROC analysis agreed closely with those of a full Bayesian analysisJournal of Clinical Epidemiology, 2004
- Approximations to the Log-Likelihood Function in the Nonlinear Mixed-Effects ModelJournal of Computational and Graphical Statistics, 1995
- A note on the application of simple linear regression methods for trend detection at multiple sites and visitsStatistics in Medicine, 1993
- Approximate Inference in Generalized Linear Mixed ModelsJournal of the American Statistical Association, 1993
- Sample Size Required for Estimating the Standard Deviation as a Per Cent of its True ValueJournal of the American Statistical Association, 1950