Bayesian approaches to meta-analysis of ROC curves

A comparative review of important classic and Bayesian approaches to fixed- effects and random-effects meta-analysis of binormal ROC curves and areas u nderneath them is presented. The ROC analyses results of seven evaluation s tudies concerning the dexamethasone suppression test provide the basis for a worked example. Particular attention is given to fully Bayesian inference , a novelty in the ROC context, based on Gibbs samples from posterior distr ibutions of hierarchical model parameters and related quantities. Fully Bay esian meta-analysis may properly account for the uncertainty associated wit h the model parameters, possibly incorporating prior knowledge and beliefs, and allows clinically intuitive predictions of unobserved study effects vi a calculation of posterior predictive densities. The effects of various dif ferent prior specifications (six noninformative as well as one informative) on the posterior estimates re investigated (sensitivity-analysis). Recomme ndations and suggestions for further research are made. Computer code for t he more advanced methods may either be downloaded via the Internet or be fo und elsewhere.