Expanding Hybrid Studies for Implementation Research: Intervention, Implementation Strategy, and Context

Successful implementation reflects the interplay between intervention, implementation strategy, and context. Hybrid effectiveness-implementation studies allow investigators to assess the effects of both intervention and implementation strategy, though the role of context as a third independent variable (IV) is incompletely specified. Our objective is to expand the hybrid typology to include mixtures of all three types of IVs: intervention, implementation strategy, and context. We propose to use I to represent the IV of intervention, IS to represent implementation strategy, and C to represent context. Primary IVs are written first and in upper case. Secondary IVs are written after a forward slash and in lower case; co-primary IVs are written after a dash and in upper case. The expanded framework specifies nine two-variable hybrid types: I/is, I-IS, IS/i, IS/c, IS-C, C/is, C/i, I-C, and I/c. We describe four in detail: I/is, IS/c, IS-C, and C/is. We also specify seven three-variable hybrid types. We argue that many studies already meet our definitions of two- or three-variable hybrids. Our proposal builds from the typology proposed by Curran et al. (1), but offers a more complete specification of hybrid study types. We need studies that measure the implementation-related effects of variations in contextual determinants, both to advance the science and to optimize intervention delivery in the real world. Prototypical implementation studies that evaluate the effectiveness of an implementation strategy, in isolation from its context, risk perpetuating the gap between evidence and practice, as they will not generate context-specific knowledge around implementation, scale-up, and de-implementation.

Abstract licence: CC BY

• Reflections on 10 years of effectiveness-implementation hybrid studies · 2022

• Promises and pitfalls in implementation science from the perspective of US-based researchers: learning from a pre-mortem · 2022

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.