Objective: The main modeling method used in survival analysis is the Cox regression model. Checking the proportionality of hazards should be an integral part of the Cox regression model. However, in most of the studies Cox regression model is used without investigating this assumption. Thus, the aim of this study was to investigate the methods which may be used in case of nonproportional hazards and show the feasibility of these models by using breast cancer data. Material and Methods: 124 patients with breast cancer were included in this study. Cox regression model and stratified Cox regression models were used to determine the prognostic factors that affect survival time of patients. Results: Treatment type does not satisfy the proportional hazard assumption. In that case, stratified Cox regression mode was more appropriate than Cox regression model. Using the stratified Cox regression model, tumor size was an important risk factor that influenced survival time of the patients. Conclusion: It was concluded that the stratified Cox regression model was more suitable for the survival data when the proportional hazard assumption did not hold.