Current practice of longitudinal follow-up frequency in outpatient clinical research is mainly based on experience, tradition, and availability of resources. Previous methods for designing follow-up times require parametric assumptions about the hazards for the event. There is a need to develop robust, easy to implement, quantitative procedures for justifying the appropriateness of follow-up frequency. Therefore, we propose a novel method to evaluate follow-up frequency by assessing the impact of ignoring interval-censoring in longitudinal studies. Specifically, we evaluate the bias in estimating hazard ratios using Cox models under various follow-up schedules. Our simulation-based procedure applies the schedules to generated data resembling the survival curve of historical data. Using this method, we evaluate the current follow-up of Parkinson's disease patients at the University of Pennsylvania Morris K. Udall Parkinson's Disease Research Center; however, the method can be applied to any research area with sufficient historical data for appropriate data generation. To allow clinical investigators to implement this method, we provide a Shiny web application.