Guidebook on Best Practices in Public Health

Step 2. Implement

This section of the guidebook helps policy makers understand the process of implementation and highlights the need for a formal implementation process to successfully transfer a best practice intervention to a target setting. It provides a general framework for understanding implementation and helps policy makers, who select and fund interventions but do usually not implement them, a guide for assessing whether implementation is sufficiently addressed in the planned transfer of the intervention. It also suggests ways in which implementation can be monitored.

The section assumes that a transferability assessment (as described in Step 1b, Section 1.3) has been completed and that the result of this assessment is that the intervention is transferable to the target setting in its original or adapted form.

2.1. Introduction to implementing an intervention

2.1.1. The need for structured implementation

Interventions to prevent or manage key public health issues are complex. As discussed in Step 1b (Section 1.3), they involve several interacting components, often target heterogeneous populations and interact with their context. They are usually delivered by people who are part of established health or social care organisations and they may be supported by technical tools, including information and communications technology. Crucially, to achieve their desired effects on relevant health outcomes, they often aim to create an interplay between their individual components to change the behaviour of health and social care providers and enhance the delivery of services, to inform the population or patient group targeted, and, ultimately, change the behaviour of citizens and patients. While research is constantly identifying many possible ways of improving population health, for instance through changing living environments and personal behaviours, these changes do not usually occur automatically to benefit populations. Many advances are not put into practice adequately, often because the will, or skill, to change behaviour are missing (Jenkins, 2003[96]).

Because such interventions rely on the interactions of many people, tools and processes, they only succeed if these interactions are co‑ordinated through careful implementation. While appropriate selection and adaptation of interventions (as discussed in Step 1a (Section 1.2) and Step 1b (Section 1.3) are prerequisites for achieving intended outcomes, these steps on their own are insufficient to ensure success. Even an appropriate and well-designed intervention can fail if it is poorly implemented. Many studies have shown that factors related to implementation affect the outcomes achieved by health promotion and prevention interventions (see, for instance, Durlak and DuPre (2008[97]) or Nielsen (2015[98])).

For the purpose of this guidebook, implementation generally refers to “the process of putting to use or integrating evidence‑based interventions within a setting” (Brownson et al., 2015, p. 305[99]). Specifically, this section guides policy makers in ensuring that interventions identified as best practice (Step 1a, Section 1.2) and suitable for transfer (Step 1b, Section 1.3) are implemented in a specified target setting. In other words, it helps ensure that the protocols and standards defined in the intervention are effectively applied in practice.

Beyond increasing the chances that new interventions achieve their desired outcomes, implementation is important for at least two other reasons. First, policy makers need to understand how an intervention was implemented to correctly interpret its effects in terms of outcomes measured. Even if a quantitative evaluation of an intervention shows that it is effective in terms of relevant health outcomes observed, policy makers can only understand causality if they know which components of the planned intervention were actually delivered and how well this was done. Again, an observed failure could result from poor implementation while success could be due to an intervention that, in practice, was very different from what was intended (Durlak and DuPre, 2008[97]). As discussed further below and in the section on preparing implementation and in Step 3 (Evaluate), an evaluation of the intervention should therefore also include implementation-related indicators. Second, an understanding of the process of implementation, and in particular the factors that contributed to success or failure of initial implementation of a new intervention, is necessary to understand whether an intervention has to be adapted further. A problem related to implementation is that interventions, even if successful in an initial test or pilot phase, are often not sustained over time and therefore fail to deliver their potential in the longer term (ibid.).

2.1.2. Conceptual approach to implementation

For the reasons discussed above, the importance of implementation cannot be overstated. At the same time, policy makers who decide that an intervention should be transferred to a target setting and who allocate funding are rarely the same people who deliver an intervention to the targeted population. Rather, policy makers usually rely on existing health and social care organisations and their staff to deliver the intervention. The ultimate effects of interventions usually depend on changes in behaviours of the citizens or patients the intervention tries to reach. There are thus many intermediary steps between policy that aims to achieve a specified outcome through selecting and funding interventions and the ultimate achievement of these outcomes, which require the concerted effort of many different people. Research on diffusion and implementation of innovations clearly shows that, even if new interventions prove effective and make intuitive sense, they are not implemented spontaneously or naturally (Brownson et al., 2015[99]; Jenkins, 2003[96]). Indeed, implementation may be the greatest challenge in successfully transferring an intervention from one setting to another.

All of this highlights the need for formal implementation processes.1 Theory and findings from implementation research can help with defining formal processes. There is a growing literature that provides a range of theories, models and frameworks that can underpin the implementation of health-related interventions in general as well as specific types of interventions (see, for example, Brownson et al. (2015[99]) and Nilsen (2015[98])). This literature sometimes suggests that using theoretical approaches will help reduce the gap between evidence and practice that is often deployed (Nilsen, 2015[98]). At the same time, critics have argued that theory is not necessarily better than common sense for guiding implementation (ibid.). It may therefore make sense to focus on a small set of key considerations in implementation without unnecessarily over-complicating the effort.

To increase the chances that an intervention achieves its desired outcomes through successful implementation, policy makers should focus on ensuring that an implementation plan exists, that it lays out the main tasks and responsibilities of people involved, and that it specifies when their tasks will be performed. This section thus conceptualises implementation in terms of “who does what, when”.

For understanding the roles of people involved in implementation and defining their specific responsibilities and tasks, implementation can be thought of as the interaction of three systems, as suggested by Wandersman et al. (2008[100]): 1. the synthesis and translation system; 2. the delivery system; and, 3. the support system. The term system refers to a set of coherently organised tasks carried out by specific people. The primary tasks of each system and the types of professionals that constitute each system are outlined in Table 2.1. For implementation to be successful, the three systems need to work together.

Table 2.1. Three interacting “systems” in implementation

Based on the interactive systems framework for dissemination and implementation

System	Primary tasks	People responsible for tasks
1. Synthesis and translation	Distil information from research and translate it into a user-friendly format for implementation in the field, for example through preparing manuals; Plan implementation and documenting the implementation protocol; Allocate resources to the delivery system and support system and ensuring that capacity is available; Design evaluation of implementation	Planners including: Policy makers; Public health specialists; Health service and operations researchers
2. Delivery	Carry out the activities that are part of the intervention; Adapt the intervention while implementation is underway while remaining faithful to the core components of the intervention (implementation fidelity); Interact with citizens/patients who participate in the intervention	Implementers, or frontline staff, including: Health and social care professionals; Public health specialists
3. Support	Provide training prior to implementation, for example through providing general training/coaching to enhance skills and motivation in the delivery system and training specific to the intervention; Provide technical assistance while implementation is underway; Supervise and monitor implementation, through evaluating fidelity while implementation is underway; Evaluate implementation	Supporters including: Public health specialists; Health service and operations researchers; IT specialists; Administrative assistants

Note: Some people involved in the implementation of an intervention might be part of several systems at the same time. For example, people who are responsible for the intervention overall, and the teams that support them, might be both, planners and supporters. In particular in small-scale interventions, some people may be planners, implementers and supporters.

Source: Adapted by the authors from Wandersman et al. (2008[100]), “Bridging the gap between prevention research and practice: The interactive systems framework for dissemination and implementation”, https://doi.org/10.1007/s10464-008-9174-z.

The implementation of interventions goes through two sequential phases: preparation of the implementation and the actual implementation itself. A core set of steps need to be completed in each of these phases (see Figure 2.1). The preparation phase of implementation is closely intertwined with, and should directly flow from, the activities in Steps 1a and 1b (Sections 1.2 and 1.3). In this first phase, people who are part of the synthesis and translation system, referred to in this section as planners, are at work to select best practice interventions based on available evidence, assess transferability and adapt it to the context of the target stetting. The preparation phase also includes activities such as writing an implementation plan, and, crucially, building capacity in the delivery and support systems. Capacity is the availability of the necessary resources so that people in the delivery system, referred to as implementers, can deliver the intervention. These implementers are then responsible for delivering the intervention as planned, while people in the support system, the supporters, provide assistance and monitoring throughout.

A particular challenge in implementation is ensuring that the implementers “stick to the plan” that has been laid out for the intervention while giving them sufficient flexibility to make adaptations. The extent to which delivery of an intervention adheres to the plan during implementation has been frequently referred to as implementation fidelity (Carroll et al., 2007[101]; Breitenstein et al., 2010[102]). Fidelity needs to be monitored and evaluated, as discussed below. There can be an inherent tension between implementation fidelity and flexibility. However, there is not necessarily a trade‑off between fidelity (i.e. sticking to the original intervention) and adaptation (to better suit the needs of the implementer context) during delivery of an intervention. Box 2.1 discusses how both can be achieved.

Box 2.1. Adaptation of interventions versus fidelity in implementation

While implementation fidelity is central to the effectiveness of an intervention, both, fidelity and adaptation, can contribute to achieving the intended outcomes of interventions (Durlak and DuPre, 2008[97]). Evidence suggests that fidelity does not usually reach 100% and that providers often replicate some parts of an intervention but modify others (Durlak and DuPre, 2008[97]). If implementers are knowledgeable about the communities in which they work and the workflows in their organisations, which is usually the case, they can be expected to be in a good position to make adaptations to an intervention during its delivery that make it more effective in the specific context. At the same time, the intervention must remain faithful to its core components that, based on theory and prior evidence, are expected to make it effective. Fidelity tends to be higher where an intervention fits well with the current mission, priorities, and existing practices at the organisation that delivers the intervention. This suggests that some adaptions are necessary during the delivery of interventions.

The key to successful implementation is distinguishing between adaptation during delivery that is desirable or necessary, to make an intervention more effective in unforeseen ways given the specific context, and undesirable adaptation that could jeopardise the core components of an intervention likely to make it effective. It is therefore important to specify, ideally during the transferability assessment discussed in Step 1b (Section 1.3) but no later than during the preparation of implementation, the components that constitute the theoretical core of an intervention and monitor how these core components are delivered or altered when the intervention is implemented. Core components should receive emphasis in terms of fidelity while less central components of the intervention can be altered to achieve a good fit with the context and environment. Keeping track of what is altered is also important for evaluating the outcomes of the intervention. Evaluation methods and indicators may need to be adapted to take into account changes of the intervention that occurred during implementation.

Source: Based on Breitenstein (2010[102]), “Implementation fidelity in community-based interventions”, https://doi.org/10.1002/nur.20373; Carroll et al. (2007[101]), “A conceptual framework for implementation fidelity”, https://doi.org/10.1186/1748-5908-2-40; and Durlak and Pre (2008[97]), “Implementation Matters: A Review of Research on the Influence of Implementation on Program Outcomes and the Factors Affecting Implementation”, https://doi.org/10.1007/s10464-008-9165-0.

The remainder of Step 2 – Implement – lays out a set of critical steps that should be covered throughout the two phases of implementation: preparing implementation and implementing. The main steps in each of the two phases are shown in Figure 2.1 and described in detail in the sections that follow.

Figure 2.1. Guidebook overview – Step 2: Implement an intervention

The content of this section is loosely based on the Quality Implementation Framework by Meyers, Durlak and Wandersman (2012[103]), findings from the CHRODIS-PLUS Joint Action (Fullaondo, Txarramendieta and De Manuel, 2018[104]; Palmer et al., 2019[105]; Stegeman et al., 2020[58]) and other findings from the literature. It is also illustrated by practical examples from prior Joint Actions, interventions reviewed by the OECD and the literature. In line with the wider OECD Framework presented in this guidebook, the section is deliberately general, to include a wide range of different public health interventions. References to more specific guidance documents for implementation of certain types of health-related interventions is in Box 2.2.

The two phases of implementation usually occur sequentially in a single implementation of an intervention. However, larger interventions might go through successive iterations of pilot implementations and scaling so that interventions are first implemented on a small scale and evaluated before a larger-scale implementation is prepared.

Box 2.2. Other resources to guide the implementation of health interventions

Various organisations and authors provide resources to guide the implementation of specific types of health-related intervention. A non-exhaustive list of resources that can guide the implementation of health promotion and non-communicable disease interventions includes:

• Manuals by WHO on population-based salt reduction strategies (WHO, 2010[106]), marketing of foods and non-alcoholic beverages to children (WHO, 2010[107]).

• WHO resources for implementing tobacco control measures (https://www.who.int/tobacco/control/en/), including those developed as part of the Framework Convention on Tobacco Control (https://fctc.who.int/coordination-platform/resources).

• Guidance resulting from the EU “CHRODIS PLUS” joint action has resulted in a number of guidance documents for the implementation of health-related interventions in general and for more specific types of interventions, including:

  • Recommendations for the implementation of health promotion good practices (Stegeman et al., 2020[58])

  • Chronic disease interventions in general (Fullaondo, Txarramendieta and De Manuel, 2018[104])

  • Interventions that promote physical activity (Barnfield, Savolainen and Lounamaa, 2020[54])

  • Interventions that manage patients with multi-morbidity (Palmer et al., 2019[105])

  • The guide for the implementation of CHRODIS PLUS recommendations and criteria to improve the quality of care for people with chronic diseases (WP7 core writing group, 2020[108]).

• Resources resulting from the EU “JANPA” joint action on prevention of overweight and obesity in children and adolescents (http://janpa-toolbox.eu/).

• Handbooks for implementation of mental health promotion interventions (Leschied, Saklofske and Flett, 2018[109]; Barry et al., 2019[110]).

2.2. Step 2a: Preparing implementation

Implementation begins with the preparation phase, which, depending on the size and complexity of the intervention, can require significant effort. Most of the critical steps of implementation occur during this phase, before actual delivery of the intervention begins (Meyers, Durlak and Wandersman, 2012[103]).

Many activities during preparation, albeit not all, are closely intertwined with the assessment of transferability and adaptation of the intervention to the context of the target stetting, as described in Step 1b, Section 1.3. While assessing transferability is concerned with understanding if the intervention can be delivered in the target setting, preparing implementation goes further than that and aims to ensure that the intervention is actually delivered. To make this happen, a central objective of the preparation phase is to build implementation capacity, in terms of human and institutional capacity, and ensuring that the attendant requirements for funding are met. Capacity for implementation is essential for professionals and organisations in the delivery system and their success or failure in implementation depends on all the factors that contribute to capacity (Durlak and DuPre, 2008[97]; Wandersman et al., 2008[100]). Another objective is to anticipate as many of the risks that could jeopardise the intervention in the specific target setting and ensure that these are addressed before they materialise.

The planners who constitute the synthesis and translation system take the leading role in this phase, such as policy makers, public health specialists and health service and operations researchers (see Table 2.1). However, implementers also have to be actively involved to ensure that the intervention can be integrated with existing processes and workflows and to take detailed account of contextual factors that may impact delivery of the intervention, and ultimately its outcomes. Planners may face a number of common challenges during preparation. Box 2.3 lists common challenges in this phase of implementation and possible approaches to meeting them.

Box 2.3. Common challenges in preparation and possible strategies to meet them

Planners are likely to encounter at least some of the following, common challenges when preparing implementation.

 

Challenge	Possible approaches to meeting this challenge
Stakeholder resistance	Engage stakeholders, including civil society Adopt a public communications strategy, including engagement of the media Create alliances with stakeholders who are supportive of the intervention Gain support from local leadership, including policy makers and managers Embed the intervention in organisations whose missions goals are aligned Initially focus implementation on people (providers and participants) who are open to change and motivated to create “early success stories” that build support Identify and foster local champions who can catalyse enthusiasm and acceptance of the intervention in existing structures
Lack of political support	Ensure intervention aligns with overarching political priorities (see Section 1.3.2, “Political context”) Ensure close collaboration and/or contact with key political leaders throughout the implementation period, including updates on progress and outcomes (if available) Identify a “champion” within the political sphere to provide ongoing support
Lack of understanding of intervention by implementers or lack of skill	Document the core components of the intervention concisely and define the scope for adaptation Document the general skills implementers are required to have and align staffing plans Organise in-depth training for all implementers to gain intervention-specific knowledge and skills Allow implementers to visit sites that have already implemented the intervention previously, to understand characteristics of the intervention and the context of prior implementations
Lack of capacity in organisations expected to implement the intervention	Prepare detailed descriptions of resources needed and tasks to be performed to implement the intervention Compare resource needs to existing capacity and identify and fill gaps Build active support by senior leaders of implementing organisations so that staff perceive the implementation as conducive to their career rather than an additional burden or career risk
Staffing shortages	Adapt intervention to make better use of existing staff Hire additional staff
Funding gaps	Adapt intervention to make better use of existing structures and resources Reduce the scope of the intervention to make it feasible with existing funding constraints Engage stakeholders to gain further political support and raise additional funds

Source: Based on Barnfield, Savolainen and Lounamaa (2020[54]), “Health Promotion Interventions: Lessons from the Transfer of Good Practices in CHRODIS-PLUS”, https://doi.org/10.3390/ijerph17041281; and Jenkins (2003[96]), “Building a better health: A handbook of behavioural change”, https://www.paho.org/hq/dmdocuments/2010/9275115907_reduce.pdf.

Preparing implementation comprises at least six key steps, which are shown in Figure 2.1 and described in detail below. These steps are presented in a logical sequence. However, they do not necessarily occur sequentially and in this order. In practice, many of the tasks might be performed in parallel. They are also interrelated. Several iterations of the various steps might be necessary, for instance to reassign roles and responsibilities, to re‑train staff or make other adjustments as new contextual information becomes available, until preparation is completed and a final implementation protocol can be written. In some cases, for instance for small and straightforward interventions or when sufficient implementation capacity is already available, certain steps might be skipped.

For many interventions, one output of the preparation phase is an implementation protocol or plan – in other words, comprehensive documentation of the remaining implementation process. The main components of an implementation protocol are spelled out in more detail below. Of course, the need for formality in implementation and the extent of such documentation increases with the size of the intervention and the number of stakeholders involved. The implementation of a small-scale intervention that targets no more than a few dozen people and is delivered by a small group of professionals that are close to the organisation that selected and funds the intervention requires less formality and less documentation than a large‑scale intervention that is delivered to hundreds or thousands of people by a broad group of professionals and involves many stakeholders. Policy makers need to make good judgment of the level of formality required.

2.2.1. Identifying people who actively contribute to implementation and all other stakeholders

The preparation phase of implementation begins with identifying all people involved in implementation and mapping out their roles. This is a task for the planners, most likely the people or organisation who selected the intervention and decided that it should be transferred in the target setting, such as a regional government or health authority, or local health care provider network. In this step, people in the synthesis and translation system also define which organisations and individual people who make up the delivery and support systems.

Agreement should be reached on which entity takes on organisational responsibility for delivery of the intervention. Results of the assessment of the political and economic context described in Step 1b (Section 1.3) can guide the identification of the right entity, if this hasn’t happened already. The entity can be the same that selected the intervention or a related entity that has a more operational role in the delivery of services related to the public health threat targeted by the intervention, such as a municipal public health department. However, care should be taken that the overall objectives of the intervention are aligned with the mission of the entity that takes on organisational responsibility so that implementation of the intervention is conducive to the achievement of the broader aims of this entity. Committed leadership is a significant enabler of successful implementation (Stegeman et al., 2020[58]).

Roles and responsibilities should be specified in this step for all people who directly contribute to implementation of the intervention. This includes all frontline professionals who deliver the intervention through interaction with the target population – the implementers – and all people who assist and support them – the supporters. See Box 2.4 for an example case study.

Box 2.4. Case study: CHRODIS‑PLUS Integrated Multimorbidity Care Model (Aragón, Spain) and Care Integration Pilot (Novo mesto, Slovenia) – roles and responsibilities and stakeholder engagement in implementation

The Department of Health of the Regional Government of Aragón took organisational responsibility for the pilot implementation of the CHRODIS‑PLUS Integrated Multimorbidity Care Model (IMCM) across 13 primary care health centres and 3 general hospitals in the region. To ensure that all people who directly contributed to implementation of the intervention were represented in preparation of the implementation, a local implementation working group (LIWG) was formed under leadership of the Department of Health. This group comprised 36 participants from different areas and institutions, including patient representatives, subject matter experts from within and outside of Aragón, and frontline staff from the primary care centres and hospitals delivering the intervention. The group collectively developed an online educational tool for health professionals participating in the intervention. Eleven members from within the group took on the responsibility for co‑ordinating implementation of the intervention following the agreed plan.

In the Novo mesto pilot to integrate care between the hospitals and community health centres, the Novo mesto General Hospital took organisational responsibility for implementation and established a LIWG that included representatives of nearly all stakeholder groups including: hospitals; community health centres; and patient representatives. Representatives from the institutions of various backgrounds joined the LIWG, including surgeons and other physicians, community and hospital nurses, social workers and administrative staff. The involvement of patient representatives in the LIWG was found helpful in finding a “common ground” and basis for communication between the various professional groups involved and ensuring that care integration was approached from the perspective of patient needs. The National Institute of Public Health, representatives of the municipality, health insurance funds and other patient organisations were identified as external stakeholders. They were engaged and encouraged to support the intervention through regular progress meetings and the organisation of community events to discuss patient experiences and barriers encountered.

Source: Gimeno Miguel et al. (2020[111]), “WP6: Pilot implementation of the CHRODIS IMCM in Aragón for improving care of patients with multimorbidity Implementation Report”, http://chrodis.eu/wp-content/uploads/2021/01/chrodis-plus-d6.1-pilot-implementation-of-integrated-care-model-for-multimorbidity-1.pdf; Piletič et al. (2020[112]), “WP7‑ Slovenia Individual pilot action report”, http://chrodis.eu/wp-content/uploads/2020/07/a-final-t7.2-slovenia.docx; and personal communications with implementers.

In addition to identifying the organisations and people who are responsible for the intervention and actively contribute to its implementation, it is helpful to identify and map all other stakeholders – i.e. people who have an interest in or are affected by the intervention, even if they do not play an active role in the implementation. A particular focus in this process lies with people and organisations who, despite not contributing actively to the implementation, are in a position to affect implementation of the intervention and its outcomes This starts in the transferability assessment discussed in Step 1b (Section 1.3), in which the “decision problem” is defined and the “political context” is assessed to gain commitment to and support of the intervention by key decision-makers, such as senior representatives of local authorities.

In this phase of implementation, however, stakeholder mapping needs to be broader and go beyond identifying the main decision-makers whose support is necessary for the intervention to succeed. Other stakeholders obviously include the target population, i.e. the patients or citizens who participate in the intervention and will be served and directly affected by it. But they can also include other government departments, local and regional authorities; industry; associations of health professionals; local universities and research institutes; the media; and NGOs, consumer groups, and other civil society groups that may be active in the field.2 Again, the need to formally identify, engage and manage stakeholders varies with the size of the intervention and the number of stakeholders involved.

Stakeholder mapping serves two main purposes. First, it helps to identify and engage with all people and organisations who can give relevant input to the implementation of the intervention. This is not only a way to adapt the intervention to the target setting but also a way of creating a sense of ownership among stakeholders, increasing the chances that implementation will succeed. Second, it helps identify all people and organisations whose interests should be taken into account in the implementation and provides a basis for managing them. The OECD provides principles and best practices for governments in consulting stakeholders, to ensure transparency and a balanced representation of interests.3 These principles can also be applied when engaging stakeholders in public health interventions. A common way of mapping stakeholders for managing them is to identify their respective interests in the intervention and their power to influence it (Ackermann and Eden, 2011[113]). This helps focusing attention on the most important actors:

• Stakeholders with a high level of power and interest require the highest level of attention. Their objectives might be aligned with those of the intervention, in which case they can be harnessed as enablers, or opposed to the objectives of the intervention, in which case they need to be engaged and managed as resistors (also see Griffiths, Maggs and George (2008[114])).

• People with low interest but high power can also be a focus of engagement, to build their awareness and support, converting them to enablers.

Various levers are available for interventions to change behaviours of stakeholders that can ultimately result in health gains. The main levers are discussed in Box 2.5.

Box 2.5. Considering the levers for changing the behaviour of stakeholders

Stakeholder engagement

Engagement of the people who will be affected by an intervention early in its conception and in implementation can, in and of itself, be a strong lever for changing behaviours and, ultimately, achieving the desired outcomes of the intervention. This is because engagement and the opportunity for people to influence and take ownership of the intervention can build self-esteem and intrinsic motivation to participate in its implementation. It also helps adapt the intervention better to the target setting. Stakeholder engagement can help consider cultural norms, openness to new ideas, and the appropriateness of the intervention in the target setting, for instance in terms of economic and social environment. Because health cannot be imposed on anyone, it is important to recognise that interventions are done with people rather than for the people they aim to reach.

Competition

The desire to emulate and outdo successful peers can be a powerful motivation for changing behaviour. Where ethically sound, this phenomenon can be a harnessed for implementing public health interventions. In particular in large‑scale interventions, planners can identify “champions” among the implementers – insiders in the group whose behaviour the intervention aims to change, who are open to change and most motivated to adopt new behaviours or ways of working. Once these people have successfully adopted the core components of the intervention, they can communicate their personal experience. Their anecdotes, enthusiasm and success can push other people to follow. Group insiders have more motivating power than outsiders. Tacit competition, and sometimes envy, among group insiders who share the same problems and culture are stronger motivators than facts and logic.

Provider payment, financial and non-financial incentives

Providing financial and non-financial incentives can facilitate implementation of an intervention. Financial incentives can be considered for people targeted by the intervention and providers delivering the intervention. For providers, non-financial incentives can also be a lever.

To increase uptake in the population or patient target group, interventions should not result in additional financial costs to participants. Direct financial rewards can be offered to people targeted by the intervention and providers delivering it, although this can be costly and may decrease intrinsic motivation. For providers, interventions often result in the shifting of tasks between organisations or individual health professionals. Planners need to understand how such task shifting may affect provider revenues, for example where revenue is generated from fee‑for-service (FFS) or case‑based payments such as diagnosis-related groups (DRGs), and whether the resulting financial incentives are conducive to or a barrier to successful implementation of the intervention. Changing remuneration schemes for providers can be a strong lever for encouraging implementation.

In addition to levers such as peer pressure and competition, a range of non-financial incentives can be considered for providers. These include, for example, offering prospects of career advancement, providing tokens of recognition for personal achievement among superiors and peers, reducing workload outside of the intervention and/or offering more flexible working hours, and changes in the organisation that create a more pleasant working environment. Such motivators may be attractive in resource constrained environments, where there is limited funding for financial incentives. However, they have to be designed carefully to be closely aligned with strategic objectives, local cultural and personal norms and values, and circumstances in provider organisations. Putting them in place effectively can therefore require a significant investment of time and energy and commitment among leadership of the organisations involved in the intervention.

Regulation

Direct regulation is one way of changing the behaviour of stakeholders that affects the outcome of an intervention. Regulation can, for example, govern the roles and responsibilities of providers who deliver an intervention, restrict the sale, use or consumption of products that are harmful to health and steer health behaviour in a population by promoting healthy and discouraging unhealthy activities. However, regulation may only be appropriate for population-based interventions, in which rules can be applied uniformly to an entire jurisdiction. Effective regulation that achieves its desired effects and avoids unintended consequences can also be difficult to design, time consuming to adopt, and costly to enforce.

Source: Based on International Council of Nurses (2008[115]), “Guidelines: Incentives for health professionals”, http://www.whpa.org/sites/default/files/2018-11/WHPA-positive_practice_environments-guidelines-EN.pdf; Jenkins (2003[96]), “Building a better health: A handbook of behavioral change”, https://www.paho.org/hq/dmdocuments/2010/9275115907_reduce.pdf; and Promberger (2013[116]), “When do financial incentives reduce intrinsic motivation? Comparing behaviors studied in psychological and economic literatures”, https://doi.org/10.1037/a0032727.

2.2.2. Setting measurable objectives and selecting relevant implementation indicators

At this stage, the planners also set appropriate objectives for implementation of the intervention in the target setting and define relevant indicators to measure progress towards and achievement of these objectives. This stage can also help understand to what extent the objectives of the intervention are aligned with the existing mission, strategy and priorities of the implementers and the organisations that make up the delivery system. Changes to the composition of the delivery system can be made if it turns out that there are significant conflicts between the objectives of the intervention and the organisational goals of the implementers.

The setting of objectives and indicators for implementation is not a stand-alone exercise. Rather, it should be done as part of designing the evaluation of the intervention and creating the logic model for evaluation, and should ideally occur before implementation (see end of Step 1 (Select) and in more detail in Step 3 (Evaluate), Section 3.2.1). Setting meaningful objectives and indicators has to take into account results of the transferability assessment in Step 1b (Section 1.3) and nearly all other decisions made in the preparation phase. While it should start early, several iterations might be necessary to complete a final version of objectives only towards the end of this phase.

Indicators for implementation are usually related to inputs and activities (see Step 3 (Evaluate), Section 3.2.2) – that is, they are part of the process evaluation that measures if the inputs and activities that make up the intervention have been deployed and delivered as planned. They can also be related to outputs and intermediate outcomes – measuring if the delivery of the intervention led to the achievement of intermediary results of the intervention. Frequently used examples of indicators to monitor and evaluate implementation are related to acceptability, adoption (also referred to as uptake), fidelity and cost (Peters et al., 2014[117]; Proctor et al., 2011[118]). Implementation-related indicators are usually not concerned with final outcomes, which are subject of the outcome evaluation that may occur some time after implementation is completed as detailed in Step 3 (Evaluate). See Box 2.6 for example case study related to indicators to measure implementation.

Box 2.6. Case study: CHRODIS‑PLUS Multimorbidity Care model – implementation-related indicators

The preparation of implementation of the Integrated Multimorbidity Care Model at the five CHRODIS‑PLUS pilot sites, for example, included an explicit step in which planners agreed on key performance indicators (KPIs) for the implementation. Activity-related indicators to monitor implementation of the interventions included, among many others, the existence of a document describing the functions/role of the case manager; the percentage of patients in the intervention for whom a case manager was identified; and, over a specified period of time, the number of visits by complex chronic patients with individual care plans to primary health care centres and the number of health care professional team meetings related to individual care plans.

Source: Palmer et al. (2019[105]), “A Methodological Approach for Implementing an Integrated Multimorbidity Care Model: Results from the Pre‑Implementation Stage of Joint Action CHRODIS-PLUS”, https://doi.org/10.3390/ijerph16245044.

When defining implementation-related indicators, it is important to keep in mind these indicators serve two distinct purposes (Saunders, Evans and Joshi, 2005[119]; Durlak and DuPre, 2008[97]):

1. To monitor implementation while it is underway, to provide feedback and make adjustments as necessary. This allows for correcting issues as implementation unfolds to increase the chances that the current implementation in the target setting achieves its objectives.

2. To evaluate implementation after it has been completed. This aims to distil and document lessons learnt so that they can be applied in future implementations. These lessons can contribute to improving the same intervention for further transfers to new target settings and future implementations but can also inform implementations of other interventions. Because of the powerful impact of implementation on outcomes discussed above, an assessment of implementation is an integral part of evaluating the intervention and allows for understanding why an intervention may have succeeded or failed.

Consideration therefore needs to be given to the points in time at which implementation-related indicators are relevant and whether it is feasible to generate them at those points in time. Some indicators are generally relevant during implementation, to monitor, provide feedback and adjust implementation as necessary. Indicators related to inputs and activities, such as the number of trained staff available, the volume of services delivered to participants in the intervention, may typically be relevant for monitoring implementation while it is underway, for example, using administrative data or observing how the intervention is delivered. Indicators that gauge short-term outcomes, such as acceptability of the intervention, might rely on surveys with professionals and patients or citizens. However, it might not always be feasible to conduct these surveys and analyse their results while implementation is underway, either because of time and/or resource constraints or because even short-term outcomes only become apparent after the implementation is completed. This may be an inherent limitation of the implementation process that cannot always be overcome. Nevertheless, planners and supporters can try to rely on informal feedback gathered from implementers of and participants in the intervention, for example through site visits and progress meetings, to gain a sense of how the intervention fares during implementation against indicators of short-term outcomes. See Box 2.7 for an example case study where implementation indicators were monitored during the implementation process.

Box 2.7. Case study: Collaborative self-management intervention promoting physical activity (Catalonia, Spain) – implementation-related indicators

In the implementation of a collaborative self-management intervention to promote physical activity in Catalonia, the planners at Hospital Clínic de Barcelona and the University of Barcelona are defining key performance indicators to be tracked during the first year of implementation. The indicators span inputs, in particular costs of the intervention to the health care system; activities that are part of the intervention, such as the number of participants enrolled compared to the number of participants targeted, and the use by participants of the personal health folder, an electronic tool that provides educational information, self-assessment questionnaires, and apps to monitor physical activity plans; and short-term outcomes of the intervention, including participant satisfaction, emergency department and general practitioner visits, and avoidable hospitalisations and hospital readmissions. KPIs are monitored during implementation. They are also analysed at the end of the first year of implementation to adjust and refine the intervention, and scale it up for regional deployment in all of Catalonia.

Source: Barberan-Garcia et al. (2018[120]), “Protocol for regional implementation of collaborative self-management services to promote physical activity“, https://doi.org/10.1186/s12913-018-3363-8; and Barberan-Garcia and Cano (2020[121]), “Assessment of the Prehabilitation Unit at Hospital Clinic of Barcelona: period 2017‑2019. NEXTCARE Implementation Report”.

2.2.3. Assessing resource needs against existing capacity and readiness

Once the people who will be the implementers and the supporters are identified, the resource needs for implementation need to be compared to existing capacity to assess the level of readiness of the existing delivery system and identify gaps. This comparison can guide capacity building – making sure that the delivery system is staffed appropriately, that professionals who are part of the delivery system have all the required skills and that the infrastructure and technical support are available. The larger the gap identified, the more attention has to be paid and effort made to build the necessary capacity.

The assessment of transferability described in Step 1b (Section 1.3) should already provide a good basis for identifying gaps between resource needs and existing capacity. In particular, assessment of the health workforce and the capital resources available in the sector specific context of the target setting (see Table 1.7) and assessment of the economic context (Table 1.9) of the target setting should address much of what needs to be done in this step.

The aim of this step is to increase the level of detail of the analysis and interact directly with the implementers to gain a complete understanding of resource needs and existing capacity. If not yet available already from the transferability assessment, a detailed “costing” exercise can help capture the resource needs of all individual tasks and processes that make up the intervention, in terms of working time of professionals and needs for physical infrastructure, equipment and technical support. The skills professionals need to deliver the intervention also have to be identified. Comparing detailed resource requirements to existing capacity can guide capacity building.

Being realistic when anticipating resources needed is a key enabler of successful implementation (see Box 2.8). Relatedly, the selection of implementation-related indicators and setting of objectives (see above) have to consider what is realistically achievable with the resources available. Experience indicates that implementations often struggle because resource needs, in particular human resources required for implementing an intervention in a new target setting, are often underestimated and objectives may be too ambitious.

Box 2.8. Case study: Anticipating resource needs in transfer of health promotion interventions in the CHRODIS‑PLUS Joint Action (JA)

Evaluation from the CHRODIS‑PLUS JA highlights the importance of being realistic when setting objectives and anticipating resource needs. Despite recognising this in the implementation framework, planners and implementers in various interventions, including JOGG (“Young People at a Healthy Weight” in Iceland), ToyBox (Malta), the Multimodal Training Intervention (Lithuania and Spain), and the Care Integration Pilot (Slovenia), reported that the time and resources required to undertake certain steps were underestimated. For example, the time needed to translate documentation and materials for participants was underestimated, there was insufficient space on existing premises and infrastructure for conducting meetings, and implementers reported a lack of support and guidance by planners. The ToyBox intervention relied on teachers to deliver the intervention to school children, who reported not receiving sufficient support from the local health promotion and disease prevention authority to enable them to deliver effective sessions. Teachers were asked to follow an implementation manual but would have benefited from more “hands‑on” support by planners. In the Care Integration Pilot in Slovenia, implementers reported that they lacked time to deliver the intervention on top of their daily work. Although key personnel were received some remuneration for additional activities that were part of the implementation, such payments did not fully cover incremental working time, leading to unpaid after hours and weekends.

Source: Piletič et al. (2020[112]), “WP7‑ Slovenia Individual pilot action report”, http://chrodis.eu/wp-content/uploads/2020/07/a-final-t7.2-slovenia.docx; Stegeman et al. (2020[58]), “D5.3 Recommendations for the implementation of health promotion good practices: Building on what works: transferring and implementing good practice to strengthen health promotion and disease prevention in Europe”, http://chrodis.eu/wp-content/uploads/2021/01/chrodis-plus-d5.3-recommendations-for-the-implementation-of-health-promotion-good-practices-1.pdf; and personal communications with implementers.

By the end of this stage, a budget for the implementation should be finalised. The assessment of resource needs and existing capacity should result in a clear idea of the incremental funding that is necessary for the delivery and support system to deliver the intervention, above and beyond the usual operating budget of the organisations that participate. If budget gaps are identified, additional funds can be raised at this stage or the implementation plan adjusted to make the implementation feasible with the budget that is available.

2.2.4. Building human and institutional capacity

Depending on the assessment of resource needs and existing capacity, significant effort in the preparation phase might need to be put towards building implementation capacity. This includes both, human capacity in the delivery system, in terms of time and skills of staff, and institutional capacity, in terms of all resources that support the people who deliver the intervention.

Building human capacity includes ensuring that sufficient staff are available, either by hiring new staff or freeing up time of existing staff members of organisations that participate in the intervention, and training the future implementers and supporters. Not surprisingly, training of staff is a factor that is strongly associated with successful implementation (Durlak and DuPre, 2008[97]). Staffing decisions should make sure that people have the necessary general and professional skills to deliver the intervention. For example, depending on the intervention, there may be a need for teams of physicians with a particular specialisation, community nurses, allied health professionals and support staff, such as information technology specialists. Intervention-specific training builds upon these existing skills to convey the knowledge and competencies necessary to deliver the intervention. If different professionals who will be implementers and supporters are expected to work in teams during implementation, the teams should be defined in this step and their members trained together. For example, in the implementation of a tele‑monitoring intervention in Denmark, training in the use of tele‑medicine equipment for all professionals in the intervention led not only to confidence in use of the equipment but also to enhanced collaboration between provider institutions due to the diversity of participants in the training sessions (Rasmussen et al., 2015[122]). Joint training sessions, or implementation planning meetings, that involve all staff that contribute to the implementation are not only key in building human capacity but also help create ownership of the intervention among staff (see Box 2.9).

Box 2.9. The importance of creating ownership in building human capacity

Experience from the transfer of good practice intervention in the CHRODIS-PLUS Joint Action (JA) suggests that it can be difficult to “transfer” complex interventions that aim to change workflows in health and social care. This is because a top-down approach to having staff implement a pre‑specified intervention that was developed for another context may result in scepticism and resistance, especially if the intervention is perceived to create incremental workload. Therefore building human capacity in preparing implementation is not only about training but also about creating intrinsic motivation and a sense of ownership of the intervention among key staff, through a recognition among them that there is a shared issue at hand that the intervention will help change. Joint training sessions can allow implementers to discuss the problem at hand and find their own solutions, even if these turn out to be similar to those already foreseen in the intervention.

Source: Personal communications with planners and implementers in interventions of the CHRODIS-PLUS JA.

At the end of training, all staff involved in the implementation should have a clear and consistent idea of why, when, where, with whom and how to implement what. They should also be reasonably self-sufficient, individually and as teams where necessary, in performing the tasks that constitute the intervention (Meyers, Durlak and Wandersman, 2012[103]).

Building institutional capacity is an essential step to establish the support system for implementation. It includes preparing all infrastructure that will support the implementation and creating an environment that is supportive of and conducive to delivering the intervention as intended. Infrastructure can include physical space, such as meeting or patient consultation rooms, digital information and communication technology (ICT), transportation, and whatever else may be immediately necessary for delivering the intervention. A conducive environment provides support and motivation. For large interventions, this can include the recruitment and training of dedicated staff who do not interact directly with the target population but support the people who deliver the intervention. These people can include public health and health service researchers, IT specialists and administrative staff. For smaller interventions, this type of support may be provided by existing staff of institutions that are involved in the implementation. In both cases, these people need to be trained to understand their role in the intervention and have the necessary time and skills to support implementation. Creating a conducive environment also comprises engagement of people who have no formal role in implementation but are present because they work in the same institution as the implementers, or in the immediate vicinity. This can take the form of informing people that are close to the intervention of its implementation, improving existing communication channels and establishing new links to stakeholders.

2.2.5. Identifying additional contextual factors to plan integration with existing processes and workflows, and define scope for further adaptation

The interaction between implementers, supporters and planners, who oversee implementation and provide intervention-specific training, is an opportunity to involve implementers in detailed planning of the implementation. This is an important step of the preparation phase for several reasons.

First, shared decision-making and active participation of the implementers in preparation can create a sense of ownership, which can in turn enhance implementation and increase the likelihood of longer-term sustainability (see Box 2.9 and Durlak and DuPre (2008[97])). At the same time, the necessary core components of an intervention should not be altered. The core of the intervention should be well established and documented by this stage. This will allow implementers to understand how flexible they can be in delivering the intervention. If any additional adaptations to the core are suggested at this stage, they need to be analysed in detail to assess whether they are necessary and might increase the likelihood of achieving desired outcomes or whether they might undermine the intervention. Any adaptations to the core should be documented and integrated with any adaptations that were made when the intervention was assessed for transferability (Step 1b (Assess transferability), Section 1.3). (The case study in Box 2.10 provides an example of how an intervention can be adapted to suit local cultural needs.)

Second, engagement with all implementers and supporters during preparation allows for identifying additional contextual factors that help appreciate how well the intervention can be integrated with existing processes and workflows. Much of this work should have already occurred by the time implementation is being prepared, as described in Step 1b (Section 1.3). But it is possible, and indeed likely, that not all contextual factors have been identified in detail upfront. Many gaps can be identified at this stage.

Third, staff in the delivery and support systems can help complete the mapping of stakeholders and the plans and strategies for engaging them as deemed necessary. This helps anticipate process dependencies and barriers to implementation. Barriers can be categorised into those that can be proactively addressed and those that are beyond the immediate control of the people who implement the intervention. Contingency plans can be formulated for the latter.

Box 2.10. Case study: StopDia for a minority population (Finland) – adapting an intervention

StopDia is a lifestyle intervention that aims to prevent type‑2 diabetes mellitus (T2DM). A small Somali community in Finland is characterised by high prevalence of T2DM and its risk factors. The intervention was adapted specifically for this Somali minority, by including members of the target group in preparation of the implementation. Translation and cultural adjustments to the intervention were completed by a Somali researcher, who also implemented the intervention through group meetings at a local mosque. The researcher was supported by volunteer students, who were also a part of the Somali community. The University of Eastern Finland, the VTT Technical Research Centre of Finland, and the Finnish Institute for Health and Welfare, who developed the StopDia intervention, provided training to the implementers and ongoing support with digital tools used in the implementation.

Source: Hussein et al. (2020[123]), “Health and wellbeing for all – development and implementation of a culturally sensitive lifestyle intervention for Somalis in Finland through the adoption of JA CHRODIS recommendations and set of criteria”; and Pihlajamäki et al. (2019[124]), “Digitally supported program for type 2 diabetes risk identification and risk reduction in real-world setting: Protocol for the StopDia model and randomized controlled trial”, https://doi.org/10.1186/s12889-019-6574-y.

2.2.6. Drafting the implementation protocol

Overall, the stages described above should result in some form of implementation plan or protocol that will guide the remaining implementation. An implementation protocol contains at least the key components listed in Table 2.2 and spells each of them out in some detail.

An implementation manual can be part of the implementation protocol or be drafted as a separate document that accompanies the protocol. While the implementation protocol guides the entity that has organisational responsibility for the implementation and includes all aspects of implementation, the manual only includes operational guidance that supports delivery of the intervention. It is a hands-on resource for implementers and supporters.

Table 2.2. Key components of the implementation protocol

Protocol component	Main content	Included in implementation manual?
Objectives	Description of the overall objective of the intervention and of the set of measurable objectives related to implementation of the intervention in the target setting	Yes
Intervention core components	Specification of a small set of activities and processes that are integral parts of the intervention and cannot be altered	Yes
Roles & responsibilities	Descriptions of specific tasks and responsibilities of all staff involved in delivering the intervention and monitoring implementation	Yes
Process maps and timelines	Detailed visualisations of relationships between staff, process workflows and related timelines for implementation	Yes
Technical assistance	Descriptions of all assistance available to staff who deliver the intervention, including support with infrastructure and equipment and any other technical assistance that is related to the intervention	Yes
Stakeholder analysis and engagement plan	Map of all stakeholders who have an interest in the intervention or can influence it and strategies to engage them as necessary	Yes, to the extent necessary for staff to understand their roles & responsibilities
Performance indicators and monitoring	Indicators monitored during implementation and description of feedback loops to ensure that indicators are communicated, discussed and acted upon	Yes, to the extent that staff and participants in the intervention have to be informed of the data collected and need information on indicators to act upon them.
Management oversight	Role of the entity that has organisational responsibility for the intervention	Yes, to the extent necessary for staff to understand their roles & responsibilities
Evaluation plan *	Indicators and data sources for final evaluation of implementation after it is completed	Yes, to the extent that staff and participants in the intervention have to be informed of the data collected. Some indicators may not be revealed until after implementation to avoid bias.

Notes: The level of detail needed in the implementation protocol and manual increases with the size and complexity of the intervention and the number of stakeholders involved. * Evaluation of the implementation might be part of the implementation protocol and/or the evaluation framework described in Step 3, Section 3.2.

2.3. Step 2b: Implementing

Following careful preparation, the most difficult part of implementation begins – ensuring that the intervention is actually delivered as planned. Much of the success or failure of implementation in this phase depends on how well implementation has been prepared and whether the people who deliver the intervention have the necessary capacity.

The implementers and supporters take the leading role in this phase (see Table 2.1). The planners, i.e. policy makers, public health specialists and health service and operations researchers, who make up the synthesis and translation system, take on the less active role of monitoring and providing assistance. This is not to understate their importance. In addition to assistance provided by the support system, ongoing monitoring as implementation unfolds can be key to success by providing supportive feedback and guiding corrective actions, identifying additional resource needs, and helping ensure that the environment in which the intervention is delivered remains conducive. Planners, implementers and supporters may face a number of common challenges during implementation. Box 2.11 lists common challenges in this phase of implementation and possible approaches to meeting them.

Box 2.11. Common challenges during implementation and possible strategies to meet them

Planners, implementers and supporters are likely to encounter at least some of the following, common challenges when implementing an intervention.

 

Challenge	Possible approaches to meeting this challenge
Loss of fidelity to the core components of the intervention	Define core components and scope for adaptation during implementation clearly during the preparation phase Monitor activity- and output-related indicators that measure implementation fidelity Provide ongoing support to implementers while implementation is underway to reinforce adherence to the core components
Limited uptake of the intervention in the targeted patient/population group	Involve representatives of the target group during the preparation phase Elicit personal characteristics of people targeted and pay special attention to people who are less likely to participate Provide clear and easy to understand information to people target about the goal and benefits of the intervention Pay attention to practical barriers to participation and organise interactions with target groups in a way that makes it as easy to participate
Insufficient time by providers to deliver the intervention as planned	Shift tasks between professionals and/or provider organisations Review tasks and remove non-essential activities Allocate additional staff to the intervention, if possible
Challenges with using technology, in particular novel digital tools that support the intervention	Provide ongoing support and re‑training to implementers with the use of technology

Source: Based on Jenkins (2003[96]), “Building a better health: A handbook of behavioral change”, https://www.paho.org/hq/dmdocuments/2010/9275115907_reduce.pdf; and de Waard et al. (2018[125]), “Barriers and facilitators to participation in a health check for cardiometabolic diseases in primary care: A systematic review”, https://doi.org/10.1177/2047487318780751.

2.3.1. Monitoring implementation and providing supportive feedback

To ensure that assistance is provided where needed most and that the necessary level of implementation fidelity is achieved, implementation should also be accompanied by monitoring and supportive feedback loops (Greenhalgh et al., 2004[126]; Fixsen et al., 2005[127]). This means that it should not only be up to implementers to raise issues and ask for support. Rather, planners or the entity that has organisational responsibility for the implementation should proactively monitor implementation by collecting and analysing data, and feeding information back to people who deliver the intervention. This will help ensure that this information is discussed and acted upon (see Box 2.12 for example case study).

Box 2.12. Case study: CHRODIS‑PLUS health promotion interventions for children and older adults (Iceland, Lithuania and Spain) – monitoring implementation

In the implementation of CHRODIS‑PLUS pilot interventions, a standard reporting template was developed and regularly filled in by implementers to report progress during implementation. The template asked short and basic questions that allowed project leaders to gain an understanding of the progress that was being made, as well as the practical barriers that implementers faced, and to provide support. Questions included:

• What did you do this month?

• What was recorded (quantitatively or qualitatively)?

• What were the successes?

• Were there any barriers?

• Was any support needed?

• What are you planning to do next month?

• What are the perceived barriers?

• How certain are you that you will achieve what you are planning to?

Source: Barnfield, Savolainen and Lounamaa (2020[54]) “Health Promotion Interventions: Lessons from the Transfer of Good Practices in CHRODIS-PLUS”, https://doi.org/10.3390/ijerph17041281.

Information from the monitoring process should be fed back in a supportive manner, to provide an opportunity for improvement and organisational or personal learning. This can either lead to making corrective actions to ensure that implementation fidelity is achieved or to adaptations of the intervention. Adapting parts of the intervention can also be a necessary response to problems that might arise during this phase. Adaptations should be made in concertation with the synthesis and translation system.

One way of providing supportive feedback to implementers is to disseminate internally any positive results observed when monitoring implementation. Dissemination refers to the active spreading of information to a target audience via determined channels (Tabak et al., 2012[128]; Yuan et al., 2010[129]). As positive results emerge, their dissemination to all implementers and other stakeholders who are actively involved in the implementation can reinforce motivation, increase fidelity to the core components of the intervention, and foster continued stakeholder support. This increases the chances that implementation is sustained through its successful completion.

Such reinforcement during implementation is particularly important for large‑scale and broad interventions, which may take a long time to implement, are dependent on the co‑operation of many implementers and on support by stakeholders. In such interventions, it is helpful to start with “quick wins” – small components of the intervention that are relatively straightforward to implement and that can show visible results quickly. Success stories about quick wins can build a sense of self-efficacy and co‑operative skills that help the groups of implementers sustain their efforts in the longer term and equips them for larger successes (Jenkins, 2003[96]).

Informal channels of communication, such as accounts of experience by implementers at meetings and exchange of staff between participating organisations, can be used. As discussed above, “champions” can be identified among implementers to disseminate results (see Box 2.3 and Box 2.5). At the same time, planners and supporters can document intermediate results and use more formal communication channels to disseminate positive results.

As more information on facilitators of and barriers to implementation of the intervention and evidence of its effectiveness emerge during the evaluation of the intervention, dissemination should also target external stakeholders. This is discussed in Step 3, Section 3.4.2.

2.3.2. Providing ongoing assistance

As implementation of the planned implementation unfolds, assistance has to be provided to the people who deliver the intervention (see Box 2.13 for example case study). Assistance while implementation is underway is one factor that is strongly related with success (Durlak and DuPre, 2008[97]). Assistance has a broad meaning here – it can include not only practical assistance with equipment, information technology and administrative processes but also any assistance related to the core components of the intervention and their delivery.

Professionals who deliver the intervention may need to ask questions about the intervention and the practical problems they encounter. They may also need a forum for discussion and peer support as unforeseen contextual factors arise. More resources than planned might turn out to be necessary to perform additional activities, tasks might have to be rescheduled, and parts of the intervention might become redundant. Assistance ensures that professionals who deliver the intervention can handle these problems.

Box 2.13. Case study: CHRODIS‑PLUS health promotion interventions for children and adults (Iceland, Lithuania and Spain) – support during implementation

During the implementation of CHRODIS-PLUS pilot interventions, the project leaders of each intervention, backed by teams of external experts and researchers, supported the implementers throughout the implementations. Project leaders made site visits, organised bi-monthly meetings via video-conference and were available for consultation by e‑mail and social media channels. Implementers were provided with tools to complete the steps defined in the implementation protocol and were involved in group discussions. Documents from these interactions, including meeting notes, recorded video-conferences and other documentation from the communication were stored and analysed to understand facilitators and barriers in the implementation process.

Source: Barnfield, Savolainen and Lounamaa (2020[54]), “Health Promotion Interventions: Lessons from the Transfer of Good Practices in CHRODIS-PLUS”, https://doi.org/10.3390/ijerph17041281.

Re‑training and activities that reinforce concepts and the skills built during training in the preparation phase are a crucial element of assistance during implementation. As professionals enter a routine, the salience of some of the concepts conveyed during training in the preparation phase might wane. Early monitoring of the implementation combined with retraining increases implementation fidelity (Durlak and DuPre, 2008[97]).

2.4. Conclusion of Step 2

Step 2 of the guidebook provides a general framework for understanding implementation in terms of “who does what, when” and is designed to provide policy makers, who may be far from the “frontline” of the delivery of an intervention that has been selected for transfer to a target setting, a practical guide to implementation. The importance of implementation cannot be overstated. As experience clearly shows, even appropriate and well-designed interventions can fail if it they are poorly implemented. Although perhaps less common, intended outcomes can conversely be due to the delivery of an intervention that, in practice, was very different from the one selected for transfer. This calls for a formal process of implementation that ensures that an intervention is delivered as planned – a concept frequently referred to as “implementation fidelity”. A clear understanding of how an intervention was implemented is also key to interpreting the results of its evaluation and acting on these results accordingly.

This step of the guidebook breaks down implementation into two sequential phases. First, an extensive phase in which implementation is prepared flows directly from the transferability assessment described in Step 1 (Select). This phase involves identifying people who actively contribute to implementation and all other stakeholders; setting measurable objectives and selecting relevant implementation indicators; assessing resource needs against existing capacity and readiness; building human and institutional capacity; identifying additional contextual factors to plan integration with existing processes and workflows, and defining the scope for further adaptation; and drafting the implementation protocol. Second, the intervention has to actually be delivered to its target group. During this phase in which implementation unfolds, this step of the guidebook suggests ways to increase the likelihood of success by monitoring implementation and providing ongoing assistance to the implementers.

Despite the importance of implementation, policy makers need to assess the level of formality required to implement an intervention, and adapt the process accordingly. The need for formal processes increases with the size of the intervention and the number of stakeholders involved. The implementation of a small-scale intervention, which is delivered by a small group of professionals that are close to the organisation that selected and funds the intervention, may require less formality than one delivered to hundreds or thousands of people by a broad group of professionals and involving many stakeholders. The general principles provided in this step of the guidebook should therefore be applied flexibly and with some common sense, taking into account the type of intervention being implemented. Policy makers can also find guidance for implementation of specific types of intervention in a number of resources listed above.

Notes

← 1.  Applying an implementation framework to guide the process is one of the recommendations of the EU Joint Action on Chronic Diseases (CHRODIS‑PLUS) report that provides recommendations for transferring and implementing good practice interventions (Stegeman et al., 2020[58]). The report also recognises that implementations science can support development of a framework, by helping analyse the context in a good practice will be implemented and how to effectively engage the people who will be affected by the implementation (ibid.).

← 2.  These are common groups of stakeholders in population-based prevention policies (see, for example, WHO (2010[106])). In general, however, one principle of stakeholder engagement is that, “all significantly affected and potentially interested parties should be consulted” (see Note 9). Which stakeholders to engage therefore depends on the type of intervention.

← 3.  See draft OECD Best Practice Principles on Stakeholder Engagement in Regulatory Policy, available at http://www.oecd.org/governance/regulatory-policy/public-consultation-best-practice-principles-on-stakeholder-engagement.htm.