Lifetime Data Analysis

, Volume 20, Issue 2, pp 303–315 | Cite as

Pseudo-observations for competing risks with covariate dependent censoring

  • Nadine Binder
  • Thomas A. Gerds
  • Per Kragh Andersen


Regression analysis for competing risks data can be based on generalized estimating equations. For the case with right censored data, pseudo-values were proposed to solve the estimating equations. In this article we investigate robustness of the pseudo-values against violation of the assumption that the probability of not being lost to follow-up (un-censored) is independent of the covariates. Modified pseudo-values are proposed which rely on a correctly specified regression model for the censoring times. Bias and efficiency of these methods are compared in a simulation study. Further illustration of the differences is obtained in an application to bone marrow transplantation data and a corresponding sensitivity analysis.


Competing risks Covariate-dependent censoring Cumulative incidence Pseudo-observations 



The research was supported by the Danish Natural Science Research Council [grant number 272-06-0442 “Point process modeling and statistical inference”]. We are grateful to CIBMTR for providing us with the example data [Public Health Service Grant/Cooperative Agreement No. U24-CA76518 from the National Cancer Institute (NCI), the National Heart, Lung and Blood Institute (NHLBI), and the National Institute of Allergy and Infectious Diseases (NIAID)].


  1. Aalen O (1978) Nonparametric estimation of partial transition probabilities in multiple decrement models. Ann Stat 6(3):534–545CrossRefMATHMathSciNetGoogle Scholar
  2. Andersen PK, Borgan Ø, Gill RD, Keiding N (1993) Statistical models based on counting processes. Springer, New YorkCrossRefMATHGoogle Scholar
  3. Andersen PK, Klein JP, Rosthøj S (2003) Generalised linear models for correlated pseudo-observations, with applications to multi-state models. Biometrika 90(1):15–27CrossRefMATHMathSciNetGoogle Scholar
  4. Andersen PK, Pohar Perme M (2010) Pseudo-observations in survival analysis. Stat Methods Med Res 19(1):71–99CrossRefMathSciNetGoogle Scholar
  5. Begun JM, Hall WJ, Huang W-M, Wellner JA (1983) Information and asymptotic efficiency in parametric–nonparametric models. Ann Stat 11(2):432–452CrossRefMATHMathSciNetGoogle Scholar
  6. Fine JP, Gray RJ (1999) A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc 94(446):496–509CrossRefMATHMathSciNetGoogle Scholar
  7. Gerds TA, Scheike TH, Andersen PK (2012) Absolute risk regression for competing risks: interpretation, link functions, and prediction. Stat Med 31(29):3921–3930CrossRefMathSciNetGoogle Scholar
  8. Gill RD (1980) Censoring and stochastic integrals. PhD thesis, Math. Centre Tracts 124. Mathematical Centre, AmsterdamGoogle Scholar
  9. Graw F, Gerds TA, Schumacher M (2009) On pseudo-values for regression analysis in competing risks models. Lifetime Data Anal 15(2):241–255CrossRefMATHMathSciNetGoogle Scholar
  10. Kalbfleisch JD, Prentice RL (2002) The statistical analysis of failure time data, 2nd edn. Wiley, New YorkCrossRefMATHGoogle Scholar
  11. Klein JP, Andersen PK (2005) Regression modeling of competing risks data based on pseudovalues of the cumulative incidence function. Biometrics 61(1):223–229CrossRefMATHMathSciNetGoogle Scholar
  12. Scheike TH, Zhang MJ, Gerds TA (2008) Predicting cumulative incidence probability by direct binomial regression. Biometrika 95:205–220CrossRefMATHMathSciNetGoogle Scholar
  13. Szydlo R, Goldman JM, Klein JP, Gale RP, Ash RC, Bach FH, Bradley BA, Casper JT, Flomenberg N, Gajewski JL et al (1997) Results of allogeneic bone marrow transplants for leukemia using donors other than HLA-identical siblings. J Clin Oncol 15(5):1767–1777Google Scholar

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  • Nadine Binder
    • 1
  • Thomas A. Gerds
    • 2
  • Per Kragh Andersen
    • 2
  1. 1.Institute of Medical Biometry and Medical InformaticsUniversity Medical Center FreiburgFreiburgGermany
  2. 2.Department of BiostatisticsUniversity of CopenhagenCopenhagen KDenmark

Personalised recommendations