|
Abstract:
|
In clinical trials and observational studies, it is often of scientific interest to evaluate the effects of covariates on complex multistate event probabilities. With continuous covariates, a common approach is to arbitrarily discretize continuous covariates, which may lead to substantial information loss, or to formulate the covariate effects in a regression model, which may have low power or be biased under misspecification. We propose nonparametric tests not requiring arbitrary discretization. General asymptotic results are derived under the null and the alternative hypotheses. The tests are consistent under stochastic ordering, which arises naturally with multistate data. Simulation studies demonstrate the gains of the new testing procedure over those based either on categorization or on regression models. The method's practical utility is illustrated with three real examples.
|
