Understanding factors affecting implementation success and sustainability of a comprehensive prevention program for cardiovascular disease in primary health care: a qualitative process evaluation study combining RE-AIM and CFIR

Abstract

Aim:

Our aim was to evaluate the implementation process of a comprehensive cardiovascular disease prevention program in general practice, to enhance understanding of influencing factors to implementation success and sustainability, and to learn how to overcome barriers.

Background:

Cardiovascular disease and its risk factors are the world’s leading cause of mortality, yet can be prevented by addressing unhealthy lifestyle behavior. Nevertheless, the transition toward a prevention-oriented primary health care remains limited. A better understanding of factors facilitating or hindering implementation success and sustainability of prevention programs, and how barriers may be addressed, is needed. This work is part of Horizon 2020 project ‘SPICES’, which aims to implement validated preventive interventions in vulnerable populations.

Methods:

We conducted a qualitative process evaluation with participatory action research approach of implementation in five general practices. Data were collected through 38 semi-structured individual and small group interviews with seven physicians, 11 nurses, one manager and one nursing assistant, conducted before, during, and after the implementation period. We applied adaptive framework analysis guided by RE-AIM Qualitative Evaluation for Systematic Translation (RE-AIM QuEST) and Consolidated Framework for Implementation Research (CFIR).

Findings:

Multiple facilitators and barriers affected reach of vulnerable target populations: adoption by primary health care providers, implementation and fidelity and intention to maintain the program into routine practice. In addition, our study revealed concrete actions, linked to implementation strategies, that can be undertaken to address identified barriers. Prioritization of prevention in general practice vision, ownership, and shared responsibility of all team members, compatibility with existing work processes and systems, expanding nurse’s roles and upskilling competence profiles, supportive financial and regulatory frameworks, and a strong community – health care link are crucial to increase implementation success and long-term maintenance of prevention programs. COVID-19 was a major barrier to the implementation. RE-AIM QuEST, CFIR, and participatory strategies are useful to guide implementation of prevention programs in primary health care.

Background

Cardiovascular diseases are the world’s leading cause of mortality with around 18,6 million deaths in 2019, representing 32% of global mortality (Nichols et al., 2014; World Health Organization, 2021) and 393 million disability-adjusted life years (Townsend et al., 2016; Roth et al., 2020). The financial burden of cardiovascular disease and its risk factors on society and the health care system is substantial (Budig & Harding, 2021). The burden is highest among individuals with lower socioeconomic status (Min et al., 2017; GBD 2017 Risk Factor Collaborators, 2018). Current evidence underpins the association between low socioeconomic status and cardiovascular disease, its risk factors, and unhealthy lifestyle behaviors (Sommer et al., 2015).

The World Health Organization estimates that nearly 75% of premature deaths are preventable (World Health Organization, 2021). Healthy lifestyle practices including smoking cessation, healthy diets, physical activity, and alcohol reduction are important in the prevention of cardiovascular disease and its modifiable risk factors such as hypertension, (pre-) diabetes, dys- and hyperlipidaemia, overweight, and obesity (Yusuf et al., 2020). Current evidence demonstrates numerous strategies to reduce cardiovascular disease risk with strong consensus on the importance of raising awareness of risk factors and on the impact of lifestyle on health outcomes (Stewart et al., 2017; Díaz-Gutiérrez et al., 2019; Hassen et al., 2021). Clinical practice guidelines yet fail to consistently propose structured protocols to guide practitioners, and gaps in evidence are reported especially regarding strategies targeting vulnerable populations (Odorico et al., 2019; Aerts et al., 2021). Consequently, people with low socioeconomic status tend to benefit less from preventive care including lifestyle interventions (Coupe et al., 2018; Rosengren et al., 2019).

A critical research–practice gap on actual implementation of structured preventive interventions indeed remains. Studies show poor achievement of guideline-recommended cardiovascular disease prevention targets (Kotseva et al., 2019, Kotseva et al., 2020). As such, there is an urgent need to further develop and implement interventions and strategies for detection and management of risk factors, in the general population as well as in vulnerable subpopulations. Horizon 2020 funded ‘Scaling-up Packages of Interventions for cardiovascular disease prevention in selected sites in Europe and sub-Saharan Africa’ (SPICES) project was established with the aim to implement evidence-based interventions for primary prevention in the population, including vulnerable groups, in low-, middle-, and high-income countries such as Belgium, where this study was carried out.

In Belgium, as in other high-income countries, prevention is primarily performed in primary health care, yet health systems fail to provide systematic support for all aspects of prevention. General practice plays a critical role in prevention and can be valuable in addressing socioeconomic health differences due to frequent contact with a large and often diverse target populations (Si et al., 2014). However, prevention-orientated services are not systematically provided in Belgian general practice. Clinical practice guidelines report various interprofessional collaboration models, including role expansion and task delegation in primary health care (Aerts et al., 2021). Integrated care delivered by physisiancs and nurses in general practice brings the opportunity to increase quality and accessibility of preventive care (Philips et al., 2015; Waller et al., 2016; Matthys et al., 2017; Srivarathan et al., 2019; Aerts et al., 2020). However, little is known about how to implement validated preventive interventions in a specific real-life context of general practice and to which extent new interdisciplinary, collaborative forms can enhance their uptake.

This study aimed to explore how a comprehensive cardiovascular disease prevention program can be implemented in general practice in a high-income country as Belgium. The aim of this study is to understand the influencing factors and facilitators for a successful implementation and sustainability and to learn how to overcome barriers. Through insight into the implementer’s experiences with the process and in the critical role of nurses, these findings provide guidance for research and practice groups that wish to scale up validated interventions for the prevention of cardiovascular disease in primary health care.

Methods

Study design and frameworks

This paper reports the qualitative process evaluation of an implementation carried out from an empowering, collaborative, and change-oriented research perspective and framed within the transformative paradigm (Creswell & Clark, 2011; Jackson et al., 2018). We applied principles of participatory action research (Koshy et al., 2011; Mortelmans, 2013) to guide the implementation process, meaning that key stakeholders were involved in the co-creation, critical reflection and dynamic, context-specific tailoring of the program throughout the different stages of our implementation study.

The expanded RE-AIM Qualitative Evaluation for Systematic Translation (RE-AIM QuEST) framework, as proposed by Forman et al. (2017), guided our formative process evaluation to identify real-time implementation barriers and explain how the context may influence sustainability and scale up to other settings (Glasgow et al., 1999; Forman et al., 2017). The complexity of the implementation context supports the use of qualitative methods as proposed by Holtrop et al., as they provide insight into ‘why and how’ our implementation process led to certain results, but it also encouraged collaborative stakeholder engagement (Holtrop et al., 2018). In this paper, we report on the qualitative evaluation of RE-AIM dimensions ‘reach’ (participation of the target population), ‘adoption’ (participation of general practices and implementers), ‘implementation’ (including fidelity), and ‘maintenance’ (of the intervention). The quantitative evaluation will be reported elsewhere, as will the qualitative evaluation of RE-AIM dimension ‘effectiveness’ from participants’ perspective.

The Consolidated Framework for Implementation Research (CFIR) (Damschroder et al., 2009; Nilsen, 2015; Nilsen & Bernhardsson, 2019), a comprehensive framework consisting of constructs associated with successful implementation, was applied to further gain understanding in implementation determinants influencing the RE-AIM dimensions. The complementary use of the RE-AIM evaluation framework and the CFIR determinant framework was previously demonstrated by King et al. (2020).

Description of the intervention and target population

The evidence-based SPICES program combines principles of Prochaska’s and Diclemente transtheoretical model (Prochaska et al., 2015), self-determination theory (Deci & Ryan, 2012), motivational interviewing (Miller & Rollnick, 2012), and brief action planning (Gutnick et al., 2014) and consists of two major components. The first ‘profiling’ component included cardiovascular disease risk stratification and communication applying the non-laboratory INTERHEART modifiable risk score (Yusuf et al., 2004). We selected this tool because of its practical usability by nurses without needing supervision or intervention of physicians. The tool uses simple questions related to lifestyle behavior risk and a waist–hip circumference measurement to allocate ones individual risk to a high-, intermediate-, or low-risk category. The profiling component was carried out by the nurse during a single session with an average duration of 20 min. The nurses used risk communication and motivational interviewing techniques to discuss the result and to initiate the appropriate follow-up trajectory. Participants either received a very brief advice on how to maintain a healthy lifestyle (low-risk score), or information on appropriate follow-up trajectories based on their individual risk score (intermediate to high-risk score).

The second ‘coaching’ component, consisted of multi-lifestyle-behavior change counseling for those at medium to high risk with one-year follow-up, spread in 10 sessions following a set interval and with a duration of approximately 30 to 45 min. The coaching sessions were focused on Dietary Approaches to Stop Hypertension (DASH) diet (Guo et al., 2021; Lari et al., 2021); combined aerobic training or aerobic and resistance physical activity; smoking cessation, and comprised behavior change techniques such as motivational interviewing, goal-setting, action-planning, and problem-solving. The intervention was delivered face to face in individual sessions. We targeted vulnerable communities using the principles of proportionate universalism (Marmot, 2012), focusing on (sub-)population-level vulnerability rather than on individual level; thus, the intended group was reached on the level of study setting. On individual level, we targeted adults between 40 and 75 years old who were not diagnosed with cardiovascular disease. People with known diabetes were excluded since they are already included in an existing care protocol including lifestyle guidance. The strategies that were used to inform, invite, and engage the target population differed in each setting, for example, passive invitation through posters; personal invitation during a contact; email or telephone invitation. Prior to the implementation, all relevant implementers received training on techniques for participant recruitment, risk profiling, and communication and lifestyle behaviour change counseling. They were also granted access to the project tools designed to support all intervention components. Both training and tools were developed by the research group in collaboration with experts in the field. A comprehensive description of the intervention, based on the Template for Intervention Description and Replication (TIDieR) checklist (Hoffmann et al., 2014), is provided as supplementary material 1, and its development will be fully described elsewhere.

Study setting and implementation

The intervention was rolled out in general practices in the Dutch-speaking Antwerp region. Multidisciplinary general practices with nurse integration were eligible for inclusion. General practices were eligible if they served a diverse population including vulnerable people with low socioeconomic status and/or if they were located in vulnerable city districts in Antwerp. Districts’ vulnerability was identified based on socioeconomic health deprivation index, limited access to primary health care, and density of households with social support. Twenty eligible practices were contacted by email or telephone, five of which were willing to participate initially (setting characteristics are summarized in Table 1). Two practices (Practice D and practice E) decided to stop participation before actual implementation took place. The three remaining settings (Practice A, B, and C) completed all implementation phases and fully implemented all intervention components. General practices did not receive any financial incentive or compensation for study participation since this would hamper sustainability of the implementation beyond the study period. They were encouraged to embed project-related activities in their regular financial system as outlined in Table 1.

Table 1. Characteristics and description of contextual factors and implementation details of included settings

Group practice: >2GPs.

* Missing data.

N/A: Not applicable.

A stepwise implementation of the intervention, developed in August 2019, was planned so that the key learnings and good practices could be scaled up from one setting to the next one. All five settings completed the pre-implementation phase, which included thorough context analysis, implementation planning, and preparation of intervention components and key implementers. In Practice A, implementation took off in September 2020; Practice B started in February 2021; and for Practice C this was in July 2021. Implementation in all settings ran until December 2021. Our stepwise approach implies a difference in duration of the per-implementation phase in each of the three implementation settings; a phase where every two to three months, we undertook reflective action research spirals, allowing the researchers and key implementers to continuously monitor the dynamic course of the implementation and to incorporate new understandings into the ongoing process. The post-implementation phase, which ran from January up till June 2022, was mainly focused on providing necessary key requirements to consolidate intervention components long term and to see how this can be scaled up to a broader context. The implementation ‘roadmap’ of each of the settings is incorporated in Table 1.

All members of the primary health care teams in each of the five included settings were considered ‘implementers’, since all of them were directly or indirectly involved in the implementation process. However, the most critical role was laid out for the nurse who carried out the intervention. The target population was approached and informed by their primary health care provider, and, if interested, they were invited to make an appointment with the nurse. In the included settings, nurses (and one nurse assistant) carried out all intervention components. In case of high-risk participants, a shared decision on the appropriate follow-up trajectory was made between nurse, physician, and participant. A total of 70 participants were profiled, 29 of which were enrolled in the coaching trajectory (Table 1).

A comprehensive analysis of the study context, including the needs and anticipated challenges to implementation, is available elsewhere (Aerts et al., 2022).

Data collection

Data collection for this process evaluation ran simultaneously to the implementation process in each setting and was completed by March 2022. Data collection primarily consisted of 38 individual or small group interviews conducted at various stages of the implementation process. Small group interviews usually consisted of two to three implementers from the same setting, providing insight into the team’s shared implementation experience through interaction. A total of 20 key implementers from the five included settings were interviewd. Key implementers were defined as implementers who were closely involved in the planning, coordination, and/or execution of the implementation and consisted of seven physicians, 11 nurses, one nursing assistant, and one practice manager. The interviews were conducted face to face when feasible, or online in video conferences depending on COVID-19-related government guidelines at the time, and each lasted between 30 and 90 min. Interviews were carried out by a team of five research assistants under the supervision of an experienced research team. All interviews were audio-recorded, and the interviewers took extensive notes during and immediately after the interviews. The interviews were transcribed as soon as possible afterward.

The main issues brought up during the interviews were regularly discussed with the larger group of implementers during their preexisting team meetings in the primary care practices. On its turn, this input was fed back to the researchers during other contact moments. This way, we ensured that the entire primary health care team in each setting was always challenged to reflection and their experience was also incorporated in our process evaluation. Additionally, we documented all implementation activities, progress, and all communications in a logbook of each setting. We kept meeting reports from all informal meetings with the implementers in order to further support thorough process mapping.

Semi-structured interview guides based on the CFIR and RE-AIM QuEST, tailored to the context and targeted implementers, were developed to answer our research questions related to each data collection phase (pre-, per-, and post-implementation) (see supplementary material 2). The topic guide included specific questions on each setting’s context, the implementation process, the facilitators and barriers to implementation of each component, adaptations that were needed, and factors influencing implementation sustainability. During this process evaluation, we also assessed the intervention components, the supporting project tools, and the implementation strategies used. In order to map the barriers and facilitators to adoption and to understand reasons for dropping out, exit interviews were also conducted with the practices that decided to drop out. We pilot-tested the interview guides and made refinements based on respondent’s feedback and researcher’s experience.

Data analysis

We analyzed all interview transcripts and documents using adaptive framework analysis (Gale et al., 2013; Ward et al., 2013) based on RE-AIM and CFIR, ensuring the possibility to also integrate text fragments that could not be placed in rigid preexisting categories. An a priori codebook was created based on RE-AIM and CFIR domains and constructs. The analysis was guided by operationalization of the four target dimensions for this study (reach, adoption, implementation, and maintenance). These clear descriptions supported the coders’ process in assigning relevant text to one of the four dimensions. Furthermore, operational definitions of CFIR domains and constructs were tailored to the study to improve coder consistency (see supplementary material 3). In the first phase of the coding process, text fragments that represented one of the four dimensions were identified. The output of the first coding phase was reviewed within the larger team of researchers, and discrepancies were solved through team discussion until consensus was reached. In a second phase, all allocated text fragments per RE-AIM dimension were subjected to a more in-depth coding procedure with the goal to further structure the text into relevant CFIR domains and (sub-) constructs. The output of this second phase was also discussed and refined based on iterative reflection cycles of the research team. Once analysis of interview data was completed, we conducted a document analysis of logbooks and meeting report guided by the final codebook. This analysis was used for the purpose of data triangulation of our primary interview data. Our data analysis was supported by QSR NVivo software version 1.5.1. This paper is built up using the consolidated criteria for reporting qualitative research (COREQ) checklist (Tong et al., 2007) and Standards for Reporting Implementation Studies (StaRI) statement (Pinnock et al., 2017) as guidance.

Results

With Table 2, we provide a comprehensive summary of facilitators and barriers to reach, adoption, and implementation, structured around relevant CFIR domains and constructs. Key findings of our per- and post-implementation process evaluation are presented in the following paragraphs structured by the covered RE-AIM domains, reinforced by implementers’ quotes. Table 3 consists of some clear examples of how barriers that were encountered by the implementers were addressed during the cyclic participatory action research process. We hereby give an overview of the implementation strategies that were applied, adapted from (Powell et al., 2015; Waltz et al., 2015), together with associated concrete actions as taken by implementers and the project team.

Table 2. Barriers and facilitators to reach, adoption, and implementation; structured by CFIR domains and constructs

Table 3. Examples of how barriers were addressed along the process; translated into implementation strategies ^* and actions related to RE-AIM dimensions

* Adapted from Powell et al. (2015) and Waltz et al. (2015) [1,2].

Reach

Personal invitation during a consultation appeared to be the best strategy to engage the target population; a strategy that was scaled up to all settings, reinforced by a poster design to inform and activate the target population.

‘For example, during our flu vaccination campaign. Most of the people we saw were eligible to participate. So we explained the project during the flu vaccination and we immediately received a lot of response’. (Nurse, Practice A)

Implementers described several factors that were taken into account when engaging people. In addition to the objective inclusion criteria, selection was also based on, for example, estimates of stage of change and the probability of effect.

‘If there are some psychological problems or they are having a hard time with something else at that moment, then I feel like that might not be the right time to open a conversation on prevention’. (Physician, Practice C)

For some of the implementers, the extent to which they felt competent also influenced the reach.

‘Certainly if they are people who have the tendency to “know better”, or already have their answer ready before you can propose something… I don’t want to coach such people, because it makes me feel so insecure. My knowledge is limited and then I come across as unprofessional’. (Nurse, Practice C)

Although active involvement of physicians in engaging the target population clearly improved reach, other priorities and insufficient involvement hindered adequate uptake of their role. The nurses developed information sheets to remind, inform, and activate the physcians as one of the actions to address this barrier.

‘I think the doctor can give some information, but I doubt if they truly familiar with all components of the project. We actually get very few patients referred. I think they just forget about it, they have a lot on their plate already during consultation’. (Nurse, Practice C)

The implementers felt the populations’s need for genuiness, authenticity, and active involvement of health care providers had become increasingly important during the COVID-19 pandemic. On the other hand, they felt that the pandemic has mainly had a negative effect on participation rates.

‘After the lockdown, we noticed that they are actually happy that they can come to us with their story, because we listen to them and show interest in their general well-being’. (Nurse, Practice B)

‘We actually see less people coming to the practice; out of fear of entering a contaminated environment … especially vulnerable people. Or fear of burdening us unnecessarily’. (Physician, Practice C)

Adoption

The implementers indicated that the implementation climate in their setting was one of the determining factors for participation in the project. There was a very strong need for a more systematic approach to prevention.

‘Prevention must absolutely improve in primary health care. That’s a fact. I think we must play a more active role in it’. (Physician, Practice E)

The project’s intervention protocols and guidance were therefore seen as a major advantage for optimizing prevention in their practice.

‘I do think the project is very valuable. It gives us the chance to specifically focus on prevention… for the first time! And it also helps that we receive support and guidance’. (Nurse, Practice A)

At the same time, implementers indicated that change is needed in the currently limited task profile of the nurse. Implementation of the intervention was therefore seen as an excellent opportunity to explore further differentiation and expansion of the professional role of nurses.

‘I think it was a good first step for the nurses to take up new tasks. They felt the need to do more than only ‘the basics’ they were doing before’. (Manager, Practice A)

‘So many protocols have been written and yet nothing has actually changed so far. While us nurses were asking for new, challenging opportunities… I actually felt a bit useless here’. (Nurse, Practice C)

The complexity and intensity of intervention components, and the associated personnel resources, were mentioned as the main barriers to adoption. This is reinforced by the lack of a financial framework for prevention and interdisciplinary collaboration from the government, which was one of the main reasons for practice E to drop out of the study since they struggled with fitting in the project activities in their regular financial system. In response to intensity as a barrier, the settings altered participant recruitment activity to the dynamics of the COVID-19 pandemic. The resources required for project- and COVID-19-related activities could not be reconciled in Practice D; the main reason why this setting has also decided to discontinue study participation.

‘Because of the time investment… I just don’t think it is feasible in this setting. And it is not only the contact with the patient, but also the burden of questionnaires and administration’. (Nurse, Practice D)

‘The government should really be encouraged to better subsidize or finance such projects. Because we have to pay for our nurses ourselves and they can’t take on other tasks during project activities’. (Physician, Practice E)

When engaging implementers, it is important that everyone is involved from the start so that the project is supported by the entire team. Moreover, it is crucial that one or more people lead the implementation within the setting, according to our respondents. The local champions and early adopters in each setting shared their experiences with the project during team meetings in order to encourage team engagement.

‘Before a practice decides whether or not to get involved, it is important that everyone knows about it, and then collectively can decide whether or not they go for it together. Of course there must be a few team members really driving through the implementation’. (Physician, Practice D)

Implementation

One of the key facilitators, mentioned by the implementers, was the adaptability of the project to each setting.

‘I think there was a lot of freedom to adapt everything to the context of our practice’. (Nurse, Practice B)

For example, the group with a high risk (red score) was also given the opportunity to participate in the coaching trajectory, after a shared decision with the nurse and physician.

‘Most people hope to get into the orange group for follow-up… they are even disappointed when they score red. So now we have decided that they can be followed up after we have consulted the doctor’. (Nurse, Practice A)

The COVID-19 pandemic was defined as one of the main barriers to the implementation.

‘We don’t know anymore… is it that we are structurally understaffed, or is it because of COVID-19. We are actually completely dependent of how the pandemic evolves, and it has a major impact on how we can plan our care and the project activities’. (Physician, Practice B)

Nurses felt that initially, physicians were not very involved, partly because the physician’s role was insufficiently clear. Implementers emphasized the importance of regular team meetings and discussion during the implementation process. Implementing the intervention has encouraged implementers to collaborate more closely in their settings, which can be facilitated by someone from the team who takes up a formal coordinating role.

‘I still miss the involvement of the doctors. I expected more feedback and more collaboration from them. I still think that the they don’t really know what is expected of them’. (Nurse, Practice A)

‘The communication in our practice has also improved as a result from implementing the intervention… because we need to discuss thinks like ’How is everything going?’ and ‘How can we do better’? We actually have to work together. We have to discuss together. We have to sit down together to see how we tackle barriers’. (Nurse, Practice C)

‘I think our practice manager has a good influence. Since she became more involved, she has proposed to bring the project on the agenda of our weekly team meeting’. (Physician, Practice A)

In all three settings, nurses have been given a more extensive and autonomous role within this prevention project. They proved to be crucial actors in the implementation.

‘I think the nurses have acquired a new role with this project. They now do part of the follow-up, which we normally did to a lesser extent. With this we were able to transfer an essential task. I think they are very suitable for this’. (Physician, Practice B)

It was seen as a major added value that nurses are able to carry out the project components from their expertise, clinical reasoning and within the medical context of a general practice.

‘We actually look beyond the profiling tool. Which makes sense, because we are trained to do so. We often measure blood pressure, or consult the patient record to see whether they take medication,… things like that’. (Nurse, Practice A)

Although they feel that the intervention machtes well with their competence profile, nurses emphasized the complexity of the coaching sessions with the aim of achieving behavioral change. After the first implementation round in Practice A, based on the nurses’ needs, the project’s training content and format were modified to increase proficiency in relevant competencies for their new role in behavior change counseling and scaled up as such in all settings.

‘Motivational interviewing… It’s difficult. I don’t really have much experience with that. With some of the participants you feel such resistance and a lack of motivation, and then I find it very difficult to get them to change their behaviour’. (Nursing assistant, Practice C)

Self-efficacy, job satisfaction, and motivation in nurses strongly depended on the results they do or do not achieve in the participants. They indicated that they needed confirmation of their abilities. In response to this need, the project team created a learning collaborative through expert-led supervision sessions where nurses received video feedback on their performance and tools for further intervision within their team.

‘I was able to give one patient a lot of information on healthy food, and he was completely open to that, while he usually is care refuser. So that went really well, and such “wins” give a lot of satisfaction’. (Nurse, Practice B)

The implementers also experienced the tension field between applying their own expertise and referring participants to community resources. The project team facilitated networking and making use of community resources, by providing assistance in navigating through the potential partnering initiatives and providers.

‘It is expected of us that we do everything ourselves. Both from the doctors and from the patients. But we aren’t specialists. We must indeed sometimes just refer people’. (Nurse, Practice C)

Maintenance

Supplementary material 4 summarizes the components that the implementers intend to sustain the intervention, as well as the end user requirements to do so, linked to relevant CFIR domains and constructs. Implementers stated that the implementation process serves as a solid basis for continuing to develop and embed the general practice-level prevention policy in the future.

‘I notice that it has triggered something in our team, … We also want to do more than providing basic care and follow-up’. (Manager, Practice A)

The need for further reflection within the team was mentioned, to outline future prevention policies and to translate and tailor good practices from the project to sustainable action plans.

‘We will have to sit down together as team to see how we are going to proceed exactly. Are we only going to focus on disease prevention or more general health promotion? How are we going to invite the patients? Which profiling tool are we going to use?’ (Nurse, Practice A)

The implementers emphasized that sustaining the project requires close follow-up and communication in order to safeguard the continuation toward common goals.

‘I think we have really learnt from this project that we need to be more responsive in the future. In the beginning there were frustrations around the project, which were left unaddressed for too long. We need to communicate about this more quickly, sit together and look for solutions’. (Physician, Practice A)

The degree of compatibility with the current system and work processes also plays a major role to what extent this will be further embedded in general practice in the future, according to the implementers.

‘Prevention is just part of our responsibility, isn’t it. We certainly try, because we have the conditions to do it here too. We work with nurses, the doctors have a very clear vision, we work with a capitation system,…’ (Nurse, Practice B)

Implementers emphasized the tension field between the relative priority of prevention compared to other core tasks of general practice, which is strongly influenced by external factors. They mentioned that reorientation toward prevention requires investment in innovative capacity building of primary health care systems.

‘The general practice is consulted for all possible problems, which makes the workflow difficult to manage… You never know what the week is going to bring, and we have especially felt it with COVID-19. We urgently need to work on resilience of the system’. (Physician, Practice B)

According to the implementers, this is also possible through role expansion of interdisciplinary team work. The nurse in particular has proven to fulfil an essential role.

‘The project proofs that primary health care is broader than the general practitioner alone. What I especially learned from that… is that you can perfectly delegate prevention to the nurses. Even better’. (Physician, Practice A)

It was mentioned that there is a need for further consolidation of nursing roles through structural and ongoing growth and strengthening their competency profile. A crucial action we undertook was to engage relevant educational institutions to respond to this need.

‘As nurses become more involved in these kinds of processes, they should receive ongoing training, for example in intervision groups with others in similar trajectories’. (Physician, Practice B)

Additionally, they stressed the importance of a strong primary health care and welfare network with care partners to rely on for certain expertise. The project team reinforced this by resonating the findings in stakeholder meetings and educational meetings with local health care and community partners.

‘We have now seen how intensive this is. It is not possible for us to acquire all that knowledge, or to offer all that in our setting. So we need a strong network actually, in the region. The practice could take on a coordinating role’. (Physician, Practice C)

Discussion

This paper describes the process evaluation of implementing a comprehensive program for the primary prevention of cardiovascular disease in five general practices in Belgium. We identified the factors that affected implementation success and sustainability and illustrated how barriers were addressed during the process by employing specific implementation strategies linked to concrete actions. Furthermore, we gained insight in the experiences of the primary health care teams with the implementation and examined nurse’s roles. These findings are meant to provide guidance for all relevant stakeholder groups that wish to scale up validated interventions for cardiovascular disease prevention in primary health care.

Several lessons have been learned during the implementation process. Foremost, the great potential of general practice as an important setting for primary prevention of cardiovascular disease, including risk profiling and lifestyle behaviour change counseling. This study especially highlighted the essential role of nurses in a transitioning primary health care toward health promotion and disease prevention and served as an opportunity to expand their scope of practice. Other studies show that nurses play a critical role in broadening, connecting, and coordinating primary and community care (Swanson et al., 2020), by applying competencies such as patient advocacy, education, and people-centered care (World Health Organization, 2020). Recent evidence states that nurses have the extensive clinical experience to deliver major improvements in primary health care (Casey et al., 2022). In various contexts, nurses increasingly and most effectively manage and coordinate care for people with, or at risk of, chronic disease, including tasks related to lifestyle risk counseling (James et al., 2019; Barr & Tsai, 2021). Despite competency potential to carry out intervention components, nurses initially felt underprepared, especially given the complex nature of behavior change interventions. Limitations of relevant competences have been previously identified as a barrier to nurses’ active involvement in preventive care (Volker et al., 2017). Our experiences are consistent with literature describing the need for ongoing education for upskilling existing nursing profiles to a more advanced level (James et al., 2019; Casey et al., 2022; Morris et al., 2022), especially with regard to patient-centered communication (James et al., 2020b), behavior change theories and counseling, and motivational interviewing; optimizing nurses’ effectiveness in communicating about lifestyle risk reduction and the reduction of chronic disease (James et al., 2020a; Hills et al., 2022). Pioneering countries in integrating nurses in general practice, such as the United Kingdom, Australia, and Canada, demonstrate that introducing quality standards, linked with quality performance reimbursement, may support ongoing professionalization, unambiguous articulation of roles and scope, and development of formal educational and career pathways, hereby enabling nurses to practice to their full scope in primary health care teams (Parker et al., 2009; Halcomb et al., 2017).

Second, this study highlights a number of barriers to reach vulnerable populations for prevention, despite the positive effects of combining engagement strategies. Reaching vulnerable populations for health promotion and prevention interventions is indeed challenging (Hoeck et al., 2014; Lim et al., 2019). When further scaling up similar preventive programs, more emphasis should be put on low-threshold approaches; population empowerment by enhancing health literacy; and social and health determinants of health care access. Our findings are supported by other research reporting on the promising context of primary health care to increase equity of health care access (Richard et al., 2016) and to decrease socioeconomic inequalities (Lorant et al., 2002; Meeus & Van Aubel, 2012).

In this study, our attempts to bolster collaborative action between general practice and community resources were limited to referral of participants to community resources, which were hindered by the lack of a strong linkage between primary health care and community organizations and lack of suitable community-led services. Our study shows the need for the currently fragmented landscape to shift toward integrated health care and welfare, by weaving networks with collaborative partnerships. In a related study within the SPICES project, which will be reported elsewhere, we also explored the opportunities of reaching vulnerable populations through existing community welfare organizations. In order to improve reach in future program planning and development, literature indeed recommends the integration of health and social care for vulnerable populations through multisectoral and community-based strategies (Richard et al., 2016; Corscadden et al., 2018). Previous studies have shown that this has great potential to increase community engagement levels and the reach of currently under-served populations, resulting in a positive impact on cardiovascular disease and its risk factors (Woringer et al., 2017; Sidebottom et al., 2021; Soltani et al., 2021).

Next, the lack of supportive financial and regulatory frameworks clarifying roles and shared responsibilities for interdisciplinary collaboration within primary health care teams were identified as main barriers to adoption. These findings are consistent with other studies describing the need for adequate funding, along with sufficient time and resources to facilitate the uptake of preventive actions in general practice and to mitigate the role constraints practitioners experience within current health systems (Volker et al., 2017; Alageel et al., 2018). Such support is also essential to enhance the continuity of preventive care and implementers’ commitment, confidence, and capacity to expand their scope of practice to systematically taking up preventive tasks (James et al., 2019; Morris et al., 2022). In accordance with our insights stipulating the structural integration of health promotion and prevention into existing work processes and systems, evidence recommends policy makers to facilitate the delivery of such interventions during routine practice (Keyworth et al., 2020). Lastly, our study revealed characteristics of the implementation setting such as networks and communications, type of collaboration, and engagement of leaders as important influencing factors to implementer commitment and fidelity. Consistent with these insights, Russell et al. emphasized the importance of tailoring preventive interventions to practice size, implementer engagement and, especially the organisation of, and relationships between, the members of the primary health care team (Russell et al., 2019).

The COVID-19 pandemic has severely impacted the implementation in terms of increased workload; focus on acute care diminishing prevention; and avoidance of unnecessary patient contacts in the context of nonurgent care and disruption of health care planning. Our experiences are in line with a study exploring the impact of the pandemic on the core competences of primary health care. They reported that preventive care was compromised and chronic care was mostly postponed and raised concerns on the profound impact of the pandemic on health, and psychological and socioeconomic wellbeing in vulnerable populations (Verhoeven et al., 2020). In addition, COVID-19 patients with preexisting noncommunicable diseases are at higher risk of severe outcomes and mortality (Alzoughool et al., 2022). Many studies during the past few years have demonstrated the negative impact of the pandemic on lifestyle behaviors related noncommunicable diseases, such as increased snacking and alcohol consumption and consequently decreased adherence to healthy diets (González-Monroy et al., 2021; Bakaloudi et al., 2021b), higher incidence of overweight and obesity (Bakaloudi et al., 2021a), and reduced physical activity and increases in sedentary time (Runacres et al., 2021). It is clear that cardiovascular disease prevention should increasingly gain the attention of primary health care providers and policy makers in order to mitigate its burden especially in vulnerable populations. We therefore argue for reprioritizing health promotion activity within primary health care systems and for shifting toward a more preventive and integrated approach (Gibson et al., 2022).

Strengths and limitations

This is the first recent study that we are aware of to combine both RE-AIM-QuEST and CFIR frameworks to examine the implementation process of a complex multicomponent intervention in real life settings in a structured and systematic way. This approach enabled us to give a comprehensive insight into key factors, set out across the different CFIR domains and constructs, that can influence the reach, adoption, implementation, and maintenance of prevention programs in primary health care. Moreover, our flexible overall study design provided ‘actionable findings’ as defined by Keith et al. (2017) and valuable information and scope for adaptations that could be made to improve the uptake into general practice, through concrete actions addressing identified barriers across the various RE-AIM domains. This study therefore provides a practical example with broad application of how the complementary use of evaluation and explanatory frameworks, nested within a participatory action research design, can explain and improve implementation success and sustainability. Our study was further strengthened by the inclusion of all key implementers of the intervention in the different settings, and by the longitudinal evaluation during the implementation process. These methods have resulted in very rich qualitative data exposing the layered effort that is required to translate evidence-based preventive interventions into daily practice. Many of our findings as well as the used methodology could be of interest to research groups, policy makers, practitioners, and all those involved in implementing related health programs in similar contexts or those tackling the challenges related to transformations in primary health care. Transferability of our findings is further reinforced by in-depth description of our study context and the rigorous use of robust implementation frameworks.

Some limitations to this study should be considered when interpreting this work. One limitation relates to the timing of the post-implementation interviews which were intended to capture information on long-term sustainability. Since we were bound to the SPICES project’s time frame and planned the interviews shortly after the implementation period, we were only able to capture the end user requirements to realize their intention of sustaining the program. Finally, this study focused solely on implementer’s perspectives. We recognize the critical importance of the views and experiences of the vulnerable target population, as evidently they are directly affected by the integration of preventive interventions of novel nature into the services provided by their trusted general practice. We did in fact include patient participants to the profiling and/or coaching components in our project evaluation, but since this called for a different methodology, we have decided to describe these findings separately.

Conclusions

The complementary use of RE-AIM QuEST and CFIR frameworks can be useful to guide the qualitative implementation process evaluation of a comprehensive intervention program for the primary prevention of cardiovascular disease in primary health care. General practice is an important setting for primary prevention of cardiovascular disease, and expanding nurse’s roles has great potential to build the capacity that is needed for scale-up and sustainability. Participatory strategies allow ongoing adaptation, enhancing uptake in practice. Actions related to adaptation to context, development of stakeholder interrelationships, and training and educating implementers are crucial to address barriers. Supportive financial and regulatory frameworks and a strong integrated community health model are needed to engage vulnerable populations and to increase long-term maintenance of prevention programs. Although COVID-19 has severely hindered implementation, our experience reinforces the urgency of health systems to shift toward a more health promotion and prevention-oriented care.