Bayesian analysis for survival of patients with gastric cancer in Iran.

BACKGROUND AND OBJECTIVES Gastric cancer is one of the most common cancers in the world. The aim of this study was to evaluate prognostic factors using Bayesian interval censoring analysis. METHODS This is a historical cohort study of 178 patients from February 2003 through January 2008, admitted with gastric cancer to one referral hospital in Tehran. Age at diagnosis, sex, histology type, tumor grade, tumor size, pathologic stage, lymph node metastasis and distant of metastasis were entered into the analysis using Bayesian Weibull and Exponential models. The term DIC was employed to find best model. RESULTS The results showed that as age increased, the risk of death slightly increased significantly in both Weibull and Exponential models with similar results. Patients with grater tumor size were also in higher risk of death followed by advanced pathologic stage. Neither the Weibull nor the Exponential models found sex, distant metastasis, histology type, tumor grade and lymph node metastasis to be prognostic factors. Based on DIC, Bayesian analysis of the Weibull model performed better than the Exponential model. CONCLUSION According to these results the early detection of patients at lower ages and in primary stages is important to increase the survival in cases with gastric cancer.