CMStatistics 2019: Start Registration
View Submission - CMStatistics
B1868
Title: Competing risks framework for repurposing of drugs Authors:  Bella Vakulenko-Lagun - University of Haifa (Israel) [presenting]
Rebecca Betensky - New York University (United States)
Sudeshna Das - Harvard Medical School (United States)
Marie-Laure Charpignon - MIT (United States)
Colin Magdamo - MGH (United States)
Yi-han Sheu - Harvard Medical School (United States)
Deborah Blacker - Harvard School of Public Health (United States)
Mark Albers - Harvard Medical School (United States)
Abstract: This simulation study is motivated by the research on repurposing of FDA-approved drugs for treating Alzheimers Disease (AD). Any research on AD has to account for competing death that might preclude the onset of AD. Both events, AD and death (before AD), are interrelated and happen commonly in elderly population, they cannot be assumed independent and should be addressed appropriately through a competing risks framework. We consider problems in estimation of the effects of static and assigned at baseline treatment and their measures of uncertainty. We compare several causal estimands, and address challenges in their relative interpretation. The simulation results provide insight, guidelines and practical recommendations for real life applications.