Modeling the association of bivariate interval-censored data under the additive hazards model

Abstract.: Bivariate interval-censored failure time data occurs in many fields such as medical followup studies. In this article, we propose an estimation procedure for the association between two related interval-censored survival times when they follow a copula model jointly and the additive hazards model marginally. A simulation study was conducted for assessing the finite sample performance of the presented method and suggests that it works well in practical situations. The proposed method is applied to the AIDS Clinical Trial Group (ACTG) 181 data that motivated this research.