A Flexible Parametric Approach to Synthetic Patients Generation in Clinical Trials

Enhancing research reproducibility and data accessibility is essential in scientific research. However, ensuring data privacy while achieving these goals is a challenging task, especially in the medical field where sensitive data are often commonplace. One solution is to use synthetic data that mimic real-world datasets. This method has the potential to streamline therapy evaluation and facilitate quicker access to innovative treatments. We propose an approach that uses sequential conditional regression and flexible parametric survival models to accurately replicate covariate patterns and survival times. The approach is easily implementable through an R function. We also provide an application to an onco-hematological trial. The results show the potentialities of the proposed method in mirroring original distributions and survival outcomes.