Department of Radiation Oncology, The Brody School of Medicine and East Carolina University, Greenville, NC 27858 USA. ahmedf@ecu.edu
The Cox proportional hazards model is the most widely used model for survival analysis because of its simplicity. The fundamental assumption in this model is the proportionality of the hazard function. When this condition is not met, other modifications or other models must be used for analysis of survival data. We illustrate in this review several methodological approaches to deal with the violation of the proportionality assumption, using survival in colon cancer as an illustrative example.