CMStatistics 2021: Start Registration
View Submission - CMStatistics
B0836
Title: Challenges and opportunities of competing risks for causal inference Authors:  Bella Vakulenko-Lagun - University of Haifa (Israel) [presenting]
Abstract: The motivation comes from research on drug repurposing for Alzheimer's Disease (AD). Any research on AD has to account for competing death that might preclude the onset of AD. We consider a problem of estimation of causal effects of a point intervention from observational data with confounding. While the guidelines for the selection of confounders in a non-competing risks framework are well established, the literature on the selection of confounders in a competing risks setting is extremely sparse - there is only one paper that addresses this problem. We employ specifics of competing risks and explore ways to adjust for confounding. Our results provide the insight and guidelines for practical causal inference applications where competing events cannot be ignored.