A0251
Title: A competing risk model for disease-specific or net survival estimation
Authors: Reuben Adatorwovor - University of Kentucky (United States) [presenting]
Abstract: Standard survival analysis methods for cause-specific survival face challenges when the cause of event information is missing or unreliable, especially in cancer registry data. To address this, we introduce a methodology for enhanced disease-specific and net survival estimation, incorporating general background population mortality data. The approach, implemented in the dMrs package, offers robust net or relative survival estimation without relying on cause-specific-event information. It accommodates diverse dependence structures and facilitates comparison of disease-specific survival under dependent competing risks. We demonstrated the effectiveness of our approach by analyzing datasets from French breast cancer and Slovenian colon cancer, resulting in improved accuracy in survival estimates and enabling robust cohort comparisons. The dMrs package provides a versatile and reliable tool for researchers, enhancing our understanding of disease-specific and net survival.
