Developments in the performance of machine learning (ML) methods have evolved significantly over the past decade with improvements in computational power (graphic processing units) combined with greatly expanded access to real world data. ML methods have traditionally focused on prediction and classification problems, but there is currently great interest in the potential of using ML for causal inference. Since regression methods have long been used for both prediction and causal inference, what are the relative merits of using ML versus regression for HEOR applications? The session will provide attendees with insight into the current state of the art in the performance of ML methods versus regression and attendees will gain valuable insight on how to apply ML methods to their own HEOR.