Simulation of left-truncated and case-k interval censored survival data with time-varying covariates

This research focuses on simulation of left-truncated and case-k interval censored survival data from the log-normal model with a time-varying covariate. Left-truncated data usually arises in prevalence cohort study where randomly selected individuals from medical records may have contracted certain disease for some duration of time but are free from event of interest at time of entry into a survival study. In this research, we proposed a simulation methodology by fixing the percentage of truncation at 20% and 60% with the width of 4 months of inspection interval. The procedure was computationally demanding due to the presence of left-truncation and time-varying covariates. The suitability of the proposed method was assessed based on the bias, standard error and root mean square of the parameter estimates for the log-normal survival model