B0829
Title: Causal inference for semi-competing risks data with application to Alzheimer's disease
Authors: Daniel Nevo - Tel Aviv University (Israel) [presenting]
Malka Gorfine - Tel Aviv University (Israel)
Abstract: The causal effects of the Apolipoprotein E4 allele (APOE) on late-onset Alzheimer's disease (AD) and death are complicated to define because AD may occur under one intervention but not under the other and because AD occurrence may affect the age of death. A semi-competing risks framework is presented to study this dual time-to-event outcome scenario. Two event times are of interest: a nonterminal event time (age at AD diagnosis) and a terminal event time (age at death). AD diagnosis time is observed only if it precedes death, which may occur before or after AD. New estimands are proposed for capturing the causal effect of APOE on AD and death. The proposal is based on a stratification of the population with respect to the order of the two events. A novel assumption is presented utilizing the time-to-event nature of the data, which is more flexible than the often-invoked monotonicity assumption. Results are derived on partial identifiability, suggest a sensitivity analysis approach, and give conditions for full identification. Finally, nonparametric and semiparametric estimation methods are presented and implemented under right-censored semi-competing risk data for studying the complex effect of APOE on AD and death.
