Background
Atrial fibrillation (AF) is one the most common heart rhythm disorder. Globally, it is estimated that 30 million individuals are diagnosed with AF [
1]. Cardiovascular morbidity and mortality related to AF remains high and is primarily associated with cardiovascular death, heart failure, stroke and hospitalisation [
1]. Individuals with AF have a three to five-fold risk of developing stroke [
2] with AF estimated to be a contributing factor of 25–30% of all ischaemic strokes [
3]. AF-related strokes are often more catastrophic and disabling than strokes of other origin [
4]. However, in the majority of cases these strokes are preventable, since in more than 25% of cases, the stroke was the first indicator of a previously unidentified AF. Prevention is two-fold, reliant on timely AF diagnosis and optimisation of thromboprophylaxis [
5].
Internationally, the proportion of undiagnosed AF is estimated to be between 11% and 30% [
3] ,which evidence an obvious need for systematic community-based screening programs. In 2016, the Atrial Fibrillation Network, the European Heart Rhythm Association, and others, have recommended the establishment of widespread screening programs for AF in individuals aged over 65, and those at high risk, including stroke survivors [
1]. To date, much attention has been directed towards promoting opportunistic AF screening and developing systematic screening programs. However, limited attention has been paid to the area of patient self-monitoring and -screening of AF, which is a suitable, yet underused practice [
6]. Given the availability of blood pressure (BP) monitoring devices that integrate reliable algorithms to detect AF one could postulate that prevention strategies may include the concept of patient self-monitoring and screening [
7‐
9]. Home BP monitoring is a widely acknowledged practice adopted by many hypertensive patients that enables multiple measurements of BP over extended periods of time [
10,
11]. Compared to an opportunistic single-screen at one point in time, self-monitoring for BP/AF at home potentially increases the chance of detecting AF. Therefore, adopting automated devices for BP monitoring that integrate reliable algorithms to detect AF is a promising practice that may enhance the diagnosis and management of AF and thus reduce AF-related cardiovascular morbidity and mortality in the long term [
9].
Community pharmacies offer a broad and accessible network of health facilities at the primary care level that provide access to a large number of individuals at risk who may be screened for a range of conditions [
12]. Additionally, there is evidence that community pharmacy services (CPSs) can positively influence patients’ behaviours, such as self-monitoring of BP [
13,
14], and, eventually, cardiovascular outcomes [
15]. To date, previous studies exploring AF-screening services in the community pharmacy have used a single, in-pharmacy opportunistic screening by a pharmacist as the principal detection method [
16,
17], yet no pharmacist-led CPSs that promote self-monitoring of AF as the main screening method have been found in the literature. Developing, implementing and evaluating novel CPSs are complex challenges [
18], which require comprehensive planning and participation of relevant stakeholders [
19,
20].
Co-design processes are participatory research approaches that engage service users, healthcare professionals and any other key stakeholders with a vested interest in a particular problem in developing (and further implementing and evaluating) health services [
21,
22]. Having stakeholders working collaboratively brings several benefits to the service planning process that can eventually facilitate their adoption into practice [
23,
24]. Generally, it fosters co-learning, nurtures networking, develops stakeholders’ positive feelings of ownership on the planned service, and brings innovative ideas, logistical and financial support. In their early stages, co-design processes require stakeholders to reflect on their own healthcare experiences, working together to identify relevant needs and priorities for improvement and to devise strategies to address the identified issues [
22]. This facilitates not only an appropriate understanding of patients’ needs and existing gaps in healthcare, but also of those circumstances that can enable or hinder the delivery and implementation of the service. As a result, co-designed health services are adjusted to the specific context or environment in which they are to be integrated, which, in turn, increases their acceptability and feasibility [
25]. Co-design processes have been used in a broad range of healthcare settings adopting diverse research approaches, generally qualitative [
26‐
29]. However, co-design has been infrequently employed in the pharmacy setting, despite the potential convenience as per existing service development and implementation challenges. Hence, the aim of this study was to use a stakeholder co-design approach to develop a pharmacist-led CPS aimed at enhancing self-management of AF (by promoting self-monitoring and screening).
Methods
A 3-step co-design process using qualitative methods (i.e., focus groups and individual interviews) was conducted to develop the aimed service. Overall, the process sequentially engaged: (1) potential service users to understand their needs and concerns and inform the next steps, (2) a mixed group of service stakeholders to work collaboratively and generate a preliminary model of the service and (3) a group of community pharmacy owners and managers to initially explore the feasibility and appropriateness of the co-designed model.
(1)
Step 1: Exploring the views of potential service users. A focus group and semi-structured interviews were conducted with individuals over 65 years with hypertension and/or AF. The focus group lasted 2.5 h and involved 4 individuals. Interviews were conducted face-to-face (
n = 2) or by phone (
n = 2). The focus group and interview guide is provided in Additional file
1. All participants were fluent in English. Individuals with a cognitive impairment, an intellectual disability or a mental illness were excluded. The aim of both the focus group and interviews was to explore: (1) the knowledge/awareness, concerns and needs of participants regarding AF, (2) their experiences with AF, hypertension (including management and monitoring) and health services; and (3) their opinions about community pharmacists and a hypothetical CPS aimed at self-screening/monitoring of AF. This latter part included specific discussions about the device Microlife BP A200 AFIB (Microlife Corp., Taipei, Taiwan), which is a clinically validated BP measurement device that integrates an algorithm for the evaluation of pulse irregularity and the detection of AF during automated BP measurement [
7‐
9]. The educational material developed by the company that has the rights to distribute the device in Australia and funded this research was assessed for appropriateness. Participants were recruited through existing networks and a not-for-profit association that organises social activities for the elderly.
(2)
Step 2: Delineating a preliminary model of the service. A 4.5-h focus group with a mixed group of stakeholders with a vested interest in the aimed CPS was conducted. Stakeholders included 8 participants who could directly or indirectly participate or have an influence in the service, including: 1 potential service user from the previous group; 2 community pharmacy owners/managers/service providers; 1 general practitioner (GP); 1 cardiologist; 1 heart failure nurse practitioner; 1 research nurse with expertise in AF; and 1 representative from the Australian Stroke Foundation. The aim of this focus group was to delineate a preliminary model of the CPS. The findings of the previous step were used to inform the group discussions, which sequentially addressed: (1) the target population of the service (i.e., Which patients would most benefit from the service?); (2) the service components, including assessments by the pharmacists, interventions targeting patients and healthcare providers, patient referral and follow-up processes (i.e., How does a high-quality pharmacy service for the targeted population look like?); and (3) what are the circumstances that can enable or hinder the delivery or implementation of the service in the existing context (i.e., what should be taken into account to integrate the service into practice?). Between the second and third part of the discussion, participants were given the educational material for assessment. The focus group guide is provided in Additional file
2. Stakeholders were purposely selected and contacted by the research team to bring a broad range of individuals and organizations that allowed for a large expertise and experience and so for designing a comprehensive service that suits current needs of patients and the healthcare system.
(3)
Step 3: Initial assessment of the feasibility and appropriateness of the co-designed model. A 2-h focus group with 4 community pharmacy owners and managers was conducted. The aim of this focus group was to further explore the feasibility and appropriateness of the CPS and assess the professional and economic implications of such a service from a business perspective. Participants were provided with an overview of the previously delineated model of the service and asked for their views about such a model. The same sequence, topics and questions delineated for the previous focus group was used (Additional file
2). Pharmacists were purposely selected and directly contacted by the research team due to their previous experience in CPS delivery.
All focus groups and interviews were facilitated by a researcher (JT) with experience in qualitative research. Focus groups took place within the university setting. Individual interviews were held in a private meeting room on campus, or were conducted by phone, according to the wishes of participants. Interviews and focus groups were audio-taped and transcribed verbatim. Ethical approval for this study was obtained by the Human Research Ethics Committee of the University of Technology, Sydney (reference number: ETH16–0276). All participants were appropriately informed about the study and written consent was obtained.
Data analysis
Data were managed in QSR NVivo (10) and analysed using Bazeley’s framework of ‘describe-compare-relate’ to identify themes and intersections between the themes [
30]. Data were analysed descriptively (i.e., categories were formed directly from participant’s responses) by one researcher (JT) and reported to the co-investigators (DSH and CF) for their comment. Subsequent analyses “challenged, extended and supported” the descriptive nodes and focused on concepts emerging from observation of all data at the more abstract level and on developing themes from these concepts. This later stage of analysis involved cycles of analysis whereby coding was reflected upon critically as new themes were developed and revised if necessary. Developing themes were compared to each other and relevant associations and differences were recorded. Themes were also compared between different focus groups and in relation to occupation and background of participants. Data from the two focus groups were analysed together in an additional analytical process. Results of this process were organised using the same structure as in the focus group: (1) potential target population, (2) service components, and (3) circumstances that may affect service implementation. To assist in the identification of these latter circumstances the list determinants of practice described by Flottorp et al... (i.e., Tailored Implementation in Chronic Disease checklist) was used [
31].
Eventually, to organise the analysed data and build a theoretical model of the service, the JeMa2 model to conceptualize pharmacy-based health programs, as described by Sabater-Hernández et al [
20], vvpharmacist interventions/pharmacy services can improve individuals’ health and quality of life by immediately affecting the determinants of both behavioural and environmental causes of health. Moreover, the model hypothesizes the relationships between pharmacy services and the context in which such services are implemented. It envisages the circumstances that can influence the implementation of the service into routine practice and the set of strategies and interventions aimed at supporting such implementation. The end result is the pharmacy-based health program comprised of the pharmacy service and the strategies and interventions that support its implementation, sustainability and overall impact [
20].
Discussion
This study summarises a 3-step co-design experience that resulted in the development of a preliminary model of a CPS aimed at encouraging self-monitoring/screening of AF. This is considered a relevant first step to suitably integrating community pharmacists with the management of AF and hence taking advantage of the accessibility of such a setting to reach a larger number of individuals in need. Importantly, the co-design process integrated the concerns, expectations and values of patients, along with the cumulated experience, knowledge and skills of key healthcare professionals, who may implement, deliver or interact with the service. As a result, it is expected the co-designed service has been adapted to suit existing needs of patients and current care practices, which, in turn, may increase the feasibility and acceptance of the service when it is implemented into a real setting.
The co-design process was predominantly driven by the self-monitoring component, which provided the original fundamentals for developing the new service and was established as the central service component from the outset. Overall, participant stakeholders agreed on the potential role of pharmacists in encouraging self-monitoring/screening practices and pursued to clearly understand how they would be adequately integrated within the pharmacy setting to enhance current AF care. Moreover, as in a previous study [
32], patients stated they would feel comfortable to self-monitor their condition and then report the results to a healthcare professional. A good example about how the self-monitoring component drove the discussions in the co-design process can be found when defining the target population. Participants stakeholders considered the target population should have an anticipated interest, motivation or need for self-monitoring and so to receive the service. Based on that argument, stakeholders prioritised individuals with existing hypertension as the target population. Also, due to the increased risk of AF of hypertensive individuals [
33], it was argued the service’s AF detection rates and thus its efficiency may be increased. Overall, the combination of both a better approach for AF screening (based on repeated measures) and a more specific target population may enhance the acceptance and feasibility of a service and overcome certain implementation challenges compared to previous, isolated AF screening programs in the community pharmacy targeting broader populations [
16,
17,
34,
35].
Patient education also emerged as a core component of the co-designed service. This is consistent with existing literature on educational needs in AF, which acknowledges patient education lays the foundation to change patient’s attitudes, behaviours and so achieve better outcomes and must address the two key challenges reported in this study (i.e., enhance awareness/knowledge about AF and manage the physiological component associated with AF) [
36]. Besides, given the self-monitoring component, patient education also becomes relevant to ensure the quality of self-monitoring [
37], which is an acknowledged international concern [
10,
11,
38]. Education in this regard must address the suitable recommendations for self-monitoring [
10,
11]. Importantly, additional educational efforts must be directed to increasing awareness and uptake of AF screening among patients for whom self-monitoring is warranted, yet they are not fully cognisant of the benefits of the practice to their individual circumstances.
In this co-design process, the JeMa2 model was used to organise the gathered information and construct the theoretical model of the CPS. The model allows for a comprehensive depiction of the CPS beyond the components of the service [
20]. Firstly, the model encompasses the set of behaviours (i.e., patients’ and physicians’ practices) and determinants of patients’ behaviours that are immediately targeted by the CPS. These intermediate variables show the ‘casual links’ that explain how the CPS is expected to improve health and inform the changes that must be assessed for the impact evaluation of the CPS [
39]. Secondly, the model addresses the context in which the CPS will be implemented, including the social, physical, economic, and policy environments. Assessing and identifying the circumstances that can positively or negatively influence implementation from the outset of the health service planning process is crucial for the early adaptation of services (i.e., adoption of best methods, procedures or materials) to best fit existing healthcare practices and also for informing the development of strategies and interventions to support the implementation of the service into practice [
25,
40].
It should be noted that despite the comprehensive approach adopted by the JeMa2 model, the model generated from this process is limited by the number of stakeholders (19 in total) who participated in an, equally restricted, number of co-design steps. Therefore, the resulted model must be seen as a preliminary approach that requires further development and decision-making before being piloted in a real setting. For example, at the service level, discussions, agreements or research are required to define a clear self-monitoring schedule for detecting AF, or patient referral or follow-up criteria. To date, guidance for recommendations to optimal method, location and frequency of AF screening are not well described within international AF guidelines. Therefore, current recommendations for BP can be used as a starting point [
10,
41]. Also at the service level, a need for conducting additional research to inform the adoption of educational methods, techniques and materials that can suit stakeholder’s needs, preferences, etc. is acknowledged. At the contextual level, the list of circumstances that may influence service implementation is not comprehensive and further groundwork and research is needed to ensure all key circumstances have been identified. Previous theory and research on determinants of healthcare practice [
31,
42] and, specifically, on barriers and facilitators to CPS implementation [
43,
44] may assist in this regard. Similarly, additional work will be necessary to identify, select and precisely define the implementation strategies and interventions once the most relevant determinants of practice have been determined [
45,
46].
Conclusions
This co-design process generated a preliminary, yet novel model of a community pharmacy-based health program that envisages the integration of community pharmacists in the management of AF; particularly, by promoting self-monitoring/screening of AF. As a result of integrating the views of multiple stakeholders, the co-designed service has been adapted to suit patients’ needs and healthcare practices, which may increase its feasibility and acceptance when implemented into practice. In future, the model may inspire pharmacy service planners and researchers to conduct further groundwork and research aimed at developing the CPS and the implementation program. Furthermore, the model may serve to prompt individual pharmacies (i.e., pharmacy owners or managers) or leading organizations (e.g., professional, scientific or governmental organizations) to make decisions for adapting the CPS to their particular context and pilot the service in a real-world setting.