Regression modelling of competing risks

Survival analysis takes into account of whether an event occurs as well as the time to such an event. Right censoring can be due to the occurrence of another event, which is called competing risks. An example of competing risk is disease progression due to a local relapse or a distant relapse. Analysis assuming censoring due to the other event can introduce bias.

In competing risks models, the probability of a local relapse is not only a hazard function of local relapses but also a hazard function of distant relapses.

This Bsc project will review the methodology for regression modelling of competing risks and develops an application of the method, and involves programming in R or other statistical software.


  • Fine J.P., Gray R.J.(1999) A proportional hazards model for the subdistribution of a competing risk. Journal of the American Statistical Association, 94(446): 496-509.
  • Satagopan J.M. et al (2004) A note on competing risks in survival analysis. British Journal of Cancer, 91, 1229-1235.

Supervisor: Dr Yinghui Wei