Our client has a platform that accelerates clinical research by helping trial sites recruit, monitor, and manage clinical trial patients.
A recent survey shows that executives running clinical trials prioritize improving trial participant retention as the most significant driver of AI investment in clinical trials.
Our client wanted to use their large collection of trial participant data to predict which participants were at risk of dropping out of studies in order to guide targeted retention strategies.
Through a combination of time series modeling, NLP, and statistics-based heuristics we created a user profile for each study participant, formalizing metrics to quantify engagement behavior.
Using these participant profiles, we trained a predictive churn model to identify behaviors that identify participants with a higher risk of dropping out of a study, enabling personalized retention efforts by study coordinators.