---

Background: Gastric cancer is the second leading cause of mortality due to cancer worldwide and it is the most common type of cancer in Iran. The aim of this study was to assess the effects of prognostic factors on the survival of patients with gastric cancer using the penalized spline in Cox model and compare it with Cox proportional hazards model. Materials and Methods: In this retrospective study, information obtained from a total of 216 gastric cancer patients who underwent surgery at Taleghani Hospital of Tehran between 2003 and 2008 years was included. Cox proportional hazards model and penalized spline in Cox model were used. R software was used for data analysis. The efficacy of these models was compared according to Akaike information criterion. Results: The five-year survival rate was 30% and the mean follow-up time was 16.60±15.28 months. Survival mean and median were 46 and 30 months, respectively. The analysis of Cox proportional hazards and penalized spline models resulted in age at diagnosis and tumor size as prognostic factors for survival time (P<0.05). Also, Akaike information criterion and survival curve for patients with a tumor size over 35 mm and age at diagnosis over 45 years were equal in both models. Conclusion: Cox proportional hazards and penalized spline models generated similar results thus, it is more appropriate to use Cox proportional hazards model because of its ease of interpretation and capability of modeling both continuous and discrete covariates. This study also showed if gastric cancer is diagnosed early, the relative risk of death will reduce.