Manu Lema Martinez

My research focuses on developing and applying advanced statistical methodologies to improve the design and analysis of clinical trials. I use Bayesian statistics and simulation to create trials that are more efficient, ethical, and robust to real-world challenges.

My previous work at Newcastle explored the theoretical framework of Bayesian adaptive designs, evaluating their performance against traditional hybrid methods. This included a deep dive into Response-Adaptive Randomization in the context of multi-arm bandit models, investigating strategies like Thompson Sampling to optimize how participants are allocated to treatments during a trial. 

My current DPhil research at Oxford applies these adaptive principles to solve practical challenges in modern, digitally-enabled platform trials in rheumatology. My focus is on developing novel statistical models to handle complex data, specifically longitudinal modeling for data with variable, patient-initiated follow-up schedules, integrating and validating digital and patient-reported outcomes (PROs) from sources like mobile apps and accelerometers.