Impact of covariate omission and categorization from the Fine–Gray model in randomized-controlled trials

In this paper, we study the statistical issues related to omission and categorization of important covariates in context Fine–Gray model randomized-controlled trials with competing risks. We show that an covariate from leads attenuated estimates for treatment effect loss proportionality general. Our simulation studies reveal substantial attenuation estimate power, while dichotomizing a continuous similar but less pronounced impact. results are illustrated using data randomized clinical trial HIV-infected individuals. The relative merits conducting adjusted versus unadjusted analysis light both practical considerations discussed.