Analysis of Survival Data under Competing Risks with Missing Cause of Death Information: Application and Implications for Study Design

Abstract

Goetghebeur and Ryan have developed a method for multiple-covariate analyses of survival data subject to competing risks when the failure type is missing for some cases. This method is compared to standard techniques over a range of missingness in failure type in a Hodgkin’s Disease data set and found to provide regression coefficients and inferences which were less biased than those from other methods.