Research

The main goal of my research is to improve statistical methods for making inferences using medical data. As a statistician engaged in clinical and epidemiological research programs, my interests reflect both medical and statistical areas. The former has been focused for the last several years on Alzheimer’s disease, breast or lymphomas cancer recurrences, and environmental health.

Statistically, the major impetus of my work has been in survival analysis. I have been studying a collection of methods that enable us to take into account correlated data in survival analysis and competing risks, with a particular interest in random effects survival models or frailty models. My other major research interest is flexible joint models for handling both recurrent events and a terminal event. These developments are associated with the development of dynamic prediction tools for predicting survival outcome.
New methods, were largely developed in the field of correlated survival data and have been shown to be far superior to more traditional methods. Nevertheless, they are still not widely used by medical scientists. I hope to change that by making these methods easier to use and better understood. This is also why I invested time in developing the R package « FRAILTYPACK » for analyzing clustered or recurrent events, or more generally correlated survival data.