Optimization of implementation strategies to maximize impact and learning

Predrag Klasnja, PhD
University of Michigan

ABSTRACT:
In recent years, behavioral science has been moving toward systematic optimization of interventions prior to their confirmatory evaluation. Intervention optimization refers to studies that are explicitly intended to improve the intervention package by assessing whether and how individual components within the package work, so that the package can be constructed only from those components that are both efficacious and that satisfy other key optimization criteria such as cost constraints. My colleagues and I have been drawing on methods from intervention optimization, as well as on Agile Science, to enable rigorous development of implementation strategies in order to more effectively target intended mechanisms of action as well as to create “usable evidence” about how and under what conditions implementation strategies are effective. In this talk I will give an overview of our approach and argue that its use in implementation science can contribute evidence about the functioning of implementation strategies that allows effective knowledge accumulation and reuse.