gms | German Medical Science

GMDS 2012: 57. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie

16. - 20.09.2012, Braunschweig

Long-term Survival for competing risk data with masked causes

Meeting Abstract

Suche in Medline nach

  • Ronny Westerman - Philipps-Universität Marburg, Institut für Medizinische Soziologie und Sozialmedizin, Marburg, Deutschland

GMDS 2012. 57. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Braunschweig, 16.-20.09.2012. Düsseldorf: German Medical Science GMS Publishing House; 2012. Doc12gmds130

doi: 10.3205/12gmds130, urn:nbn:de:0183-12gmds1303

Veröffentlicht: 13. September 2012

© 2012 Westerman.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen ( Er darf vervielf&aauml;ltigt, verbreitet und &oauml;ffentlich zug&aauml;nglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.



Competing Risks Models have a various field of applications in medical and public health studies. A challenging clue for applying cause-specific survival models yield on the problem of missing and misclassification in cause of death. The masked cause of death is related to incomplete or only partial identifiable in formation of death certificates. Different Bayesian approaches [1] e.g. the mixture cure model [2] are proposed to account for that problem. Another question is related to adequate estimates for long-term Survival in respect to the limitation of lifetime among all risks. As a new parametric distribution the long-term exponential geometric distribution (LEG) introduced by Roman et al. 2012 [3] can be considered. The LEG will be defined as three-parameters lifetime distribution with decreasing hazard function with mixed model characteristics. One of challencing remarks is to predict the conditional failure time function for future events or items depending on times of a few early failures.

The main purpose of this work is to compare the LEG with alternative parametric versions like Weibull distribution, or the simple Exponential distribution for long-term survival estimates. Data analysis will be realized with e.g. UK Stroke Register Data and R Statistical Software. As on remarkable conclusion one would expect the best fitting of the LEG for the long-term survival regarding to Weibull and Exponential distribution.


Sen A, Banerjee M, Li Y, Noone AM. A Bayesian approach to competing risks analysis with masked cause of death. Stat Med. 2010;29:1681-95.
Basu S, Tiwari RC. Breast Cancer survival, competing risks and mixture cure model: a Bayesian analysis. J Royal Stat Soc. 2010;173(2):307-29.
Roman M, Louzada F, Cancho VG, Leite JG. A New Long-Term Survival Distribution for Cancer Data. J Data Sci. 2012;10:241-58.