Get access

Concordance measures in shared frailty models: application to clustered data in cancer prognosis

Authors

  • Audrey Mauguen,

    Corresponding author
    1. Univ. Bordeaux ISPED, Centre INSERM U897-Epidémiologie-Biostatistique, F-33000 Bordeaux, France
    2. INSERM, ISPED, Centre INSERM U897-Epidémiologie-Biostatistique, F-33000 Bordeaux, France
    • Correspondence to: Audrey Mauguen, INSERM U897 - Equipe de biostatistique, ISPED, Université Bordeaux Segalen, 146 rue Leo Saignat, 33076 BORDEAUX CEDEX, France.

      E-mail: audrey.mauguen@isped.u-bordeaux2.fr

    Search for more papers by this author
  • Sandra Collette,

    1. EORTC HeadQuarters, B-1200 Bruxelles, Belgique
    Search for more papers by this author
  • Jean-Pierre Pignon,

    1. Service de Biostatistique et d'Epidémiologie, Institut de Cancérologie Gustave-Roussy, F-94805 Villejuif, France
    Search for more papers by this author
  • Virginie Rondeau

    1. Univ. Bordeaux ISPED, Centre INSERM U897-Epidémiologie-Biostatistique, F-33000 Bordeaux, France
    2. INSERM, ISPED, Centre INSERM U897-Epidémiologie-Biostatistique, F-33000 Bordeaux, France
    Search for more papers by this author

Abstract

Frailty models are gaining interest in prognostic studies, especially because of the spread of multicenter studies. However, little research has been performed to extend prognostic tools to frailty models, including discrimination measures. As previously performed for the Harrell's c-index, we extended two different discrimination measures (the model-based concordance probability estimation of Gönen and Heller and the nonparametric Uno's c-index) to take into account cluster membership. We calculate measures at three levels: between-group, where only patients with different frailties are compared, within-group, where only patients sharing the same frailty are compared, and overall. We performed simulations to study the impact of group size and the number of groups on these measures. Results showed that the two measures can be extended to frailty models while remaining independent from censoring distribution, provided that the group size is sufficient. We apply the extended measures to two real datasets, a meta-analysis and a large multicenter trial. Copyright © 2013 John Wiley & Sons, Ltd.

Ancillary