In this talk, I will provide an overview of the why, what and how of implementation science, with particular focus on statistical methods used. Next, I will provide an overview of the goals of our new center, CMIPS (Center for Methods on Implementation and Prevention Science). Then, I will discuss some of the new methods developed by my group motivated by the particular features of implementation and prevention science research and practice, including methods for causal inference of individual and disseminated effects in network-randomized studies, causal mediation analysis, stepped wedge designs, two-stage designs for cluster randomized studies, learn as you go designs, and estimation and inference for the incremental cost effectiveness ratio. Finally, I will conclude by identifying opportunities for novel methodologic development in which students and faculty at Yale can engage.