Although various definitions of implementation exist, the one presented by Damshroder and Hagedorn (2011) is used here: "Implementation refers to efforts designed to get evidence-based programs or practices of known dimensions into use via effective change strategies" (p. 195). Extensive experience indicates that when evidence-based programs are attempted by a new organization, in a new setting, or by new staff, they are not automatically reproduced or replicated with the quality intended by the program developers. For a variety of reasons, major changes can occur, so that the new program may not be an accurate reproduction of the core components of the original version.
The gap between how a program is intended by its designers to be delivered and its actual delivery in practice is referred to as implementation variation. Implementation may vary from strict adherence to program protocols as designed to subtle or major changes in program protocols. The challenge is to implement a program with sufficient quality to obtain the outcomes found in original trials. In other words, implementation exists along a continuum and one can think of poor, medium, or high quality implementation. The emphasis here is on high quality because implementation to this degree increases the chances of obtaining the outcomes found in original trials.