Implementation science comprises the study of methods to understand factors associated with successful integration of evidence-based interventions into community practice. There is evidence suggesting that it takes an average of 17 years for autism evidence-based interventions (EBI) to enter community practice settings. EBIs for children with autism often include complex and resource-intensive strategies that address different clinical outcomes that must fit with the resources and mission of the organizations in which they are implemented, and with the training and abilities of the practitioners using them. Research on use of EBIs in community settings must focus on implementation outcomes. These include the extent to which EBIs are: 1) implemented in the way they were designed (fidelity); 2) integrated within a service system (penetration); and 3) sustained once initial implementation efforts end (sustainment). We will present data from a: 1) randomized school-based implementation trial of a social engagement EBI for children with autism; 2) randomized school-based effectiveness trial of autism behavioral EBIs in public schools; 3) large-scale effectiveness trial of an autism behavioral EBI in publicly-funded mental health agencies; and 4) prospective, observational study of multiple EBIs implemented in a large-scale mental health system reform in which children with autism were served.