부산대학교 도서관

Contemporary clinical trials often have complex logistics: they are run across multiple centers/countries and involve a lot of uncertainties about disease prevalence rates, patient characteristics and regulatory requirements, all of which can vary across countries. As a result, planning and delivering such studies on-time has long been recognized as a challenge in the pharmaceutical industry. One well-known approach to modeling recruitment in complex trials, which we use as a starting point of our work is a Poisson-Gamma stochastic model. COVID-19 pandemic brought additional layer of complexity to recruitment modeling: as many countries went into shutdown, sponsors were forced with the dilemma of replanning study delivery rebalancing their portfolios "on the fly." That translated into the need to adapt the recruitment model to account for effect of COVID-19. In this work we present an approach that blends Poisson–Gamma model, real-time recruitment data updates and epidemiological modeling of COVID-19 spread. At the core of the methodology is sophisticated Bayesian hierarchical model predicting how pandemic intensity affected parameters of Poisson-gamma model. We also build an app conveniently showing many publicly available COVID models in one place. The work is a result of intense collaboration among many statisticians, data scientists and clinical operations professionals. [ABSTRACT FROM AUTHOR]