Statistical Methods for Evaluation of Biomarkers

Eunice Kennedy Shriver National Institute Of Child Health & Human Development

Investigators

Linked publications & trials

Abstract

In the past year investigators and staff on this project developed nonparametric sequential methods for evaluation of diagnostic biomarkers. They investigated a two-stage procedures for selecting the best diagnostic biomarkers to control the probability of incorrect selection. Methods for point and interval estimation of accuracies of a binary diagnostic biomarkers were developed following group sequential evaluation of the biomarkers' sensitivity and specificity. For biomarkers whose measurements are subject to limits of detection, the group developed maximum likelihood ratio tests for comparing the discriminatory ability of biomarkers subject to limit of detection. To design comparative diagnostic accuracy studies an adaptive approach was developed to maintain the study power due to misspecifications of nuisance parameters.