The superiority of Bayesian Method in the Analysis of 8-Year Survival of Breast Cancer and its Determinants in Yazd

Abstract

Introduction: Breast cancer is one of the most common diseases among women and various factors are involved in its development. The aim of this study was to determine the factors affecting the survival of women with breast cancer in Yazd using Cox's model using the Bayesian and Classic methods.

Methods: A population-based study was conducted on 538 women with breast cancer who referred to Ramezanzade Radiotherapy Center. The required data were collected from the April 2005 to March 2012. This analytical study was conducted using survival analysis method. Data were analyzed by R software version 0/4/3 and the significance level was set at 0.05.

Results: We used the Kaplan method and found that the 1-, 5-, and 8-year survival of women with breast cancer were 0.976, 0.898, 0.823, and 0.737, respectively. The mean age of participants was 48.03±11.16 and the mean survival time was about 97.64±4.23 months. Bayesian Cox regression results showed that markers of surgery (HR=1.631 95%PI; 1.102-2.422), ki67 (HR= 3.260. 95%PI; 1.6308-6.372), stage (HR=5.620, 95%PI; 4.079-7.731),   lymph node (HR= 1.765, 95%PI; 1.127-2.790), and ER (HR = 2. 600 95%PI; 2.023-3.354) were significantly related to survival.

Conclusion: Due to the short probability interval for the risk ratio, the Cox Bayesian model was selected as the optimal model. Accordingly, the variables of disease stage, lymph node involvement, the type of surgery, and markers of Ki67 and ER had positive effects on the death risk.  In this study we used the findings of the previous studies regarding the validity of the Bayesian method. So, application of this model in survival analysis requires more detailed investigations.