Meta-analysis of ROC curves

The authors present a method to combine several independent studies of the same (continuous or semiquantitative) diagnostic test, where each study rep orts a complete ROC curve; a plot of the true-positive rate or sensitivity against the false-positive rate or one minus the specificity. The result of the analysis is a pooled ROC curve, with a confidence band, as opposed to earlier proposals that result in a pooled area under the ROC curve. The ana lysis is based on a two-parameter model for the ROC curve that can be estim ated for each individual curve. The parameters are then pooled with a bivar iate random-effects meta-analytic method. and a curve can be drawn from the pooled parameters. The authors propose to use a model that specifies a lin ear relation between the logistic transformations of sensitivity and one mi nus specificity. Specifically, they define V = In(sensitivity/(1 - sensitiv ity)) and U = In((1 - specificity)/specificity), and then D = V - U, S = V + U. The model is defined as D = alpha + beta S. The parameters alpha and b eta are estimated using weighted linear regression with bootstrapping to ge t the standard errors, or using maximum likelihood. The authors show how th e procedure works with continuous test data and with categorical test data.