Common risk difference test and interval estimation of risk difference for stratified bilateral correlated data.

Bilateral correlated data are often encountered in medical researches such as ophthalmologic (or otolaryngologic) studies, in which each unit contributes information from paired organs to the data analysis, and the measurements from such paired organs are generally highly correlated. Various statistical methods have been developed to tackle intra-class correlation on bilateral correlated data analysis. In practice, it is very important to adjust the effect of confounder on statistical inferences, since either ignoring the intra-class correlation or confounding effect may lead to biased results. In this article, we propose three approaches for testing common risk difference for stratified bilateral correlated data under the assumption of equal correlation. Five confidence intervals of common difference of two proportions are derived. The performance of the proposed test methods and confidence interval estimations is evaluated by Monte Carlo simulations. The simulation results show that the score test statistic outperforms other statistics in the sense that the former has robust type I error rates with high powers. The score confidence interval induced from the score test statistic performs satisfactorily in terms of coverage probabilities with reasonable interval widths. A real data set from an otolaryngologic study is used to illustrate the proposed methodologies.

Comparing k cumulative incidence functions through resampling methods.

Tests for the equality of k cumulative incidence functions in a competing risks model are proposed. Test statistics are based on a vector of processes related to the cumulative incidence functions. Since their asymptotic distributions appear very complicated and depend on the underlying distribution of the data, two resampling techniques, namely the well-known bootstrap method and the so-called random symmetrization method, are used to approximate the critical values of the tests. Without making any assumptions on the nature of dependence between the risks, the tests allow one to compare k risks simultaneously for k > or = 2 under the random censorship model. Tests against ordered alternatives are also considered. Simulation studies indicate that the proposed tests perform very well with moderate sample size. A real application to cancer mortality data is given.