In a case-cohort design, covariate histories are measured only on cases and a subcohort that is randomly selected from the entire cohort. This design has been widely used in large epidemiologic studies, especially when the exposures of interest are expensive to assemble for all the subjects. In this paper, we propose statistical procedures for analyzing case-cohort sampled current status data under the additive hazards model. Asymptotical properties of the proposed estimator are described and we suggest a resampling method to estimate the variances. Simulation studies show that the proposed method works well for finite sample sizes, and one data set is analyzed for illustrative purposes.
In practice, it often occurs that some covariates of interest are not measured because of various reasons, but there may exist some auxiliary information available. In this case, an issue of interest is how to make use of the available auxiliary information for statistical analysis. This paper discusses statistical inference problems in the context of current status data arising from an additive hazards model with auxiliary covariates. An empirical log-likelihood ratio statistic for the regression parameter vector is defined and its limiting distribution is shown to be a standard chi-squared distribution. A profile empirical log-likelihood ratio statistic for a sub-vector of the parameters and its asymptotic distribution are also studied. To assess the finite sample performance of the proposed methods, simulation studies are implemented and simulation results show that the methods work well.