Survival Analysis of Breast Cancer Data Using Non- homogeneous Multi-State Model

The Multi-State Model can be used to determine the natural trend of disease progress. The aim of this study was to use Multi-State Model in analyzing breast cancer data.
In this descriptive-longitudinal study, data of 573 women with breast cancer were studied. Sample patients referred to Esfahan Sayed o Shohada Hospital between 1999 and 2006 and were followed up until April 2015. Data was analyzed using R 3.2.2 software.
The mean and standard deviation of women age was 47.2±10.8 years. The transition probability from first treatment state to recuperate state was 25%. The transition hazard ratio from first treatment state to recuperate state in a 5 year period time was 9. The sojourn time was 3.85 years for the first treatment, 1.11 years for recuperate, and 0.04 years for the death.
Based on transition probabilities and also survival mean in each state, the physicians could suggest the most suitable supportive care and treatment for patients especially during the first state of treatment.