Table 1 ImpRes domains and rationale underpinning inclusion

Domain 1: Implementation research characteristics

It has been argued that the research-to-practice gap is inflated by the tendency to conduct research in a step-wise manner from clinical efficacy research, followed by clinical effectiveness research, and lastly implementation research [18,19,20,21]. The benefits of simultaneously researching effectiveness and implementation to promote the more rapid and successful translation of clinical interventions have been previously described and a hybrid effectiveness-implementation typology proposed [22]. As such, the ImpRes tool encourages researchers to differentiate between studying the effectiveness of an evidence-based intervention (e.g. to improve a specific service or patient outcome) and studying the implementation of an intervention in a real-world setting (e.g. to identify barriers and facilitators to implementation and evaluate implementation success and strategies). The ImpRes tool urges researchers to categorize the degree of focus placed on evaluating the effectiveness of the intervention and/or implementation success (e.g. from effectiveness-implementation hybrid studies to pure implementation research [22]). The ImpRes tool and guide are not intended to be overly prescriptive in terms of implementation study design; different designs lend themselves to different project aims and objectives, as such there is not a one size fits all.

Domain 2: Implementation theories, frameworks and models

The use of implementation theories, frameworks and models is important in implementation research for multiple reasons including guiding implementation efforts, improving understanding of implementation determinants and providing a structure to synthesize findings [9]. Furthermore, the use of theories and frameworks enhances the generalizability of implementation efforts and help to build a cumulative understanding of the nature of change [23] enabling greater replication of evidence-based interventions in other settings. The underuse, superficial use and misuse of theories and frameworks have been described as posing a substantial scientific challenge for implementation science [9]. ImpRes does not endorse the use of any particular theory, framework or model as the aims and objectives of any given implementation research project will determine the appropriateness and relevance of particular implementation theories, frameworks and models. Depending on the aims and objectives of an implementation research project, it may be appropriate to apply more than one theory, framework or model. For example, if a project aims to identify and understand barriers and enablers to individual behavior change and simultaneously evaluate implementation efforts, the application of a determinant framework which focuses on individual level determinants, such as the theoretical domains framework (TDF) [24], as well as a framework, such as RE-AIM [25], that can be used to evaluate implementation efforts, would be appropriate. As such, the ImpRes tool encourages researchers to consider and apply implementation theories, frameworks or models in implementation research that are appropriate to the aims and objectives of the research.

Domain 3: Determinants of implementation

Implementation success is, in part, a function of the context in which implementation efforts occur. Without a clear understanding of the contextual factors likely to impede or facilitate implementation efforts, implementation strategies to overcome barriers and maximize facilitators will not be optimized, thereby reducing the likelihood of implementation success. The ImpRes tool specifically highlights the consolidated framework for implementation research (CFIR) [26], because CFIR brings together key constructs, derived from multiple implementation theories, that influence implementation efforts and is one of the most highly cited implementation determinant frameworks in the field [27]. Whilst the ImpRes tool highlights CFIR, it is important to note that ImpRes does not endorse the use of a particular determinant framework. Rather than simply describing the context in which implementation efforts occur, the ImpRes tool prompts research teams to design research to prospectively and systematically explore the factors that are likely to hinder or facilitate implementation efforts—including identification of factors commonly present across contexts that impede or facilitate health intervention implementation. Recent efforts to improve our understanding of implementation in given contexts have provided a more comprehensive, unified and structured conceptualization of context, extending beyond simply barriers and facilitators to implementation [28,29,30].

Domain 4: Implementation strategies

Described as constituting the ‘active ingredient’ and ‘how to’ components of implementation efforts, implementation strategies are regarded as having unparalleled importance in implementation science [31]. A program of research is currently underway to improve guidance regarding how best to select implementation strategies [8, 32, 33]. The ImpRes tool and guide aim to advise research teams to prospectively appraise the context in which they intend to implement evidence-based interventions and subsequently select contextually appropriate implementation strategies to maximize implementation success. The ImpRes tool and guide thus aim to facilitate explicit mapping of implementation strategies to prospectively identified barriers and facilitators to implementation efforts by signposting researchers to resources and tools designed to help researchers achieve this.

Domain 5: Service and patient outcomes

Researchers working in the field of applied health research will be familiar with, and understand the importance of, identifying, assessing and measuring relevant service and patient outcomes (commonly referred to as ‘client outcomes’ in the USA) to determine the efficacy and effectiveness of interventions. Unless conducting ‘pure’ implementation research [22], implementation outcomes [34], will be evaluated in addition to service and patient outcomes. The ImpRes tool and guide aim to highlight the limitations of only assessing service and patient outcomes when trying to understand and evaluate implementation. By making explicit the different types of outcomes an implementation study can assess, the ImpRes tool and guide facilitate the thinking that patient and service level outcomes are conceptually distinct from implementation outcomes—thus the need to assess the latter becomes prominent. In doing so, the ImpRes tool and guide aim to increase awareness that whilst the assessment and measurement of service and patient outcomes are essential to determine the clinical efficacy and effectiveness of an intervention, they are not sufficient for understanding implementation success, or failure of an intervention. For example, an intervention designed to reduce Emergency Department admissions might be found to be ineffective according to admission data (i.e. service outcomes). However, the intervention might subsequently be found to be effective, for example, when implemented as intended (i.e. implemented with fidelity).

Domain 6: Implementation outcomes

Implementation outcomes have been defined as ‘the effects of deliberate and purposive actions to implement new treatments, practices and services and are distinct from service and client (patient) outcomes.’ [34]. They may focus on a specific level of implementation, such as system, organization, innovation, provider and patient. Proctor et al. proposed a working taxonomy of eight implementation outcomes that may be assessed across all levels of a healthcare system, these include acceptability, appropriateness, feasibility, adoption, fidelity, penetration, implementation cost and sustainability [34]. Implementation outcomes may be assessed using validated measurement instruments, routinely collected data and qualitative methods. They should be assessed at salient stages of implementation (e.g. early for adoption, ongoing for penetration, late for sustainability). Furthermore, implementation strategies should be appropriately aligned to the implementation outcome(s) they intended to target and improve.

Domain 7: Economic evaluation

Implementation actions and strategies will inevitably consume scarce resources that could be put to other valued uses. Decision makers are therefore likely to want to gain insight into which options offer the greatest returns, in terms of patient or population health-related benefits, per $ of resource invested [35]. Economic evaluation provides a methodological framework for examining this question, and for adding further insight into the inter-dependencies between activities, cost, implementation outcomes and patient outcomes [36]. While the concept and importance of costing and consideration to cost-effectiveness in relation to implementation strategies is not particularly novel, to the best of our knowledge, the ImpRes tool and guide is the only implementation research tool of its type to provide guidance on data requirements for costing implementation strategies and how economic evaluation more generally should fit into a broader implementation evaluation plan.

Domain 8: Stakeholder involvement and engagement

The ImpRes tool explicitly urges researchers to view implementation research as a genuinely collaborative undertaking between researchers and stakeholders and therefore stresses the importance of involving and engaging stakeholders in the process of designing implementation research. Stakeholders are individuals and groups who are likely to be impacted or affected by implementation efforts; they may also be influential in determining the success or failure of implementation efforts (e.g. health service leaders or managers or commissioners). Involving and engaging stakeholders as part of the study design team helps to ensure that, for example, suitable implementation strategies and outcomes (both patient, service and implementation outcomes) are measured; and that an appropriate communication and dissemination plan are developed.

Domain 9: Patient and public involvement and engagement

Patient and public involvement (PPI) is widely considered a marker of high-quality research [37]. Evidence suggests that PPI in implementation research is lagging and has not matured to the same extent as PPI in other areas of health research and as such remains a largely untapped resource [38, 39]. The ImpRes tool prompts research teams to distinguish and consider PPI in isolation to other stakeholder involvement because of its critical importance in facilitating the implementation of research evidence into clinical practice and thus maximizing patient benefit and reducing health inequalities. As such, research teams are encouraged to view implementation research as a partnership between researchers and patients and thus involve patients and the public in designing implementation research.

Domain 10: Unintended consequences

Implementation efforts are often complex, challenging, and unpredictable. Unintended consequences (both positive and negative) are likely to occur as a result of implementation efforts; however, unintended consequences of implementation efforts are often not studied and/or under-reported. Research teams often focus their efforts on quantifying the intended and anticipated benefits of implementation efforts and fail to consider unintended and unanticipated consequences. The ImpRes tool prompts research teams to be mindful of and explore the potential unintended consequences of implementation efforts.