Bayesian multivariate hierarchical transformation models for ROC analysis

Authors

  • A. James O'Malley,

    Corresponding author
    1. Department of Health Care Policy, Harvard Medical School, Boston, MA 02115, U.S.A.
    • Department of Health Care Policy, Harvard Medical School, 180 Longwood Avenue, Boston, MA 02115, U.S.A.
    Search for more papers by this author
  • Kelly H. Zou

    1. Department of Health Care Policy, Harvard Medical School, Boston, MA 02115, U.S.A.
    2. Department of Radiology, Brigham and Women's Hospital, Boston, MA 02115, U.S.A.
    Search for more papers by this author

Abstract

A Bayesian multivariate hierarchical transformation model (BMHTM) is developed for receiver operating characteristic (ROC) curve analysis based on clustered continuous diagnostic outcome data with covariates. Two special features of this model are that it incorporates non-linear monotone transformations of the outcomes and that multiple correlated outcomes may be analysed. The mean, variance, and transformation components are all modelled parametrically, enabling a wide range of inferences. The general framework is illustrated by focusing on two problems: (1) analysis of the diagnostic accuracy of a covariate-dependent univariate test outcome requiring a Box–Cox transformation within each cluster to map the test outcomes to a common family of distributions; (2) development of an optimal composite diagnostic test using multivariate clustered outcome data. In the second problem, the composite test is estimated using discriminant function analysis and compared to the test derived from logistic regression analysis where the gold standard is a binary outcome. The proposed methodology is illustrated on prostate cancer biopsy data from a multi-centre clinical trial. Copyright © 2005 John Wiley & Sons, Ltd.

Ancillary