Cox regression models with nonproportional hazards applied to lung cancer survival data

The Cox regression model, which is widely used for the analysis of treatment and prognostic effects with censored survival data, makes the assumption of constant hazard ratio. In the violation of this assumption, different methods should be used to deal with non-proportionality of hazards. In this study, the stratified Cox regression model and extended Cox regression model, which uses time dependent covariate terms with fixed functions of time are discussed. The results are illustrated by an analysis of lung cancer data in order to compare these methods with respect to Cox regression model in the presence of nonproportional hazards.