My program of research is centred around improving the safety and effectiveness of medications for individuals living with multimorbidity across the continuum of care using real-world data.  It is important to me that the direction of my research is informed by unmet patient and physician needs related to drug safety. Therefore, I aim to apply advanced quantitative methods (from epidemiology, biostatistics and machine learning) to clinical problems which impact “real world” patients in every day clinical practice. Some of the quantitative methods that I use in my research include clinical predictive modelling, advanced survival analysis, unsupervised learning and reinforcement learning. I also focus on applying methods from predictive modelling within a causal inference framework.

My research is unique because it spans multiple disciplines including clinical practice, epidemiology, biostatistics, causal inference, computing science and health informatics. I regularly engage with a variety of researchers and healthcare professionals in the conduct of my research such as biostatisticians, data scientists, pharmacists, physicians, hospital administrators, software developers and IT specialist. Importantly, I also engage with patients and caregivers which is key to ensuring my research takes a patient centred approach. Therefore, I can effectively bridge the gap between advanced methods, clinical practice and patient care. 

My research also considers multiple sources of data across different countries including clinical hospital data and administrative health data in Canada, clinical data from the United Kingdom and Dutch electronic health data.