Atrial fibrillation and primary care prevention and rehabilitation – a feasibility study Authors: Caroline M Elnegaard, Maria K Pedersen, Ann-Dorthe O Zwisler, Britt Borregaard, Jesper F Eilsø, Tove Sommer, Claus Tveskov, Jeroen M Hendriks, Axel Brandes, Signe S Risom Published: 2025 ABSTRACT AbstractBackgroundThe increasing prevalence of atrial fibrillation (AF) poses a global healthcare challenge, necessitating effective management strategies. AF is associated with reduced health-related quality of life (HRQoL), higher mental distress, and frequent hospital readmissions. Interventions like patient education and physical exercise have shown promise in reducing these burdens, yet their integration into primary care in a complex intervention set-up lacks thorough qualification and evidence.AimThis study aimed to investigate the feasibility of delivering a comprehensive prevention and rehabilitation intervention for individuals with AF in a primary Healthcare Center. Secondary objectives explored participant experiences and acceptability, needs and preferences, and changes in HRQoL, anxiety, and depression as well as feasibility of collecting patient-reported outcome measures (PROMs).Trial designA single-arm, intervention, feasibility study was conducted.MethodsThe primary feasibility outcomes were to investigate acceptability among participants and impact on the key outcomes HRQoL, anxiety and depression. Participants were consecutively recruited from a university hospital outpatient AF-clinic and referred to a primary Healthcare Center. The complex intervention included patient education, physical exercise, Medicinal yoga, psychosocial support and risk factor management. Quantitative data were collected using PROMs for HRQoL, anxiety and depression, while qualitative data were obtained through focus group interviews and field notes. Data analysis involved descriptive statistics and thematic analysis in a mixed-methods evaluation.ResultsOf the 50 participants enrolled, 43 (86%) completed the intervention. The complex intervention was feasible in a primary Healthcare Center with high acceptability. HRQoL improved from baseline to end-intervention and maintained at two-month follow-up, and symptoms of anxiety and depression decreased. Qualitative findings highlighted the importance of support from healthcare professionals and peers, alongside increased knowledge and tools to manage AF symptoms.HarmsNo adverse events were reported.ConclusionsThe study demonstrated the feasibility of a complex prevention and rehabilitation intervention for individuals with AF in a primary care setting and valuable insights were gained into the potential effectiveness and acceptability of the intervention. These findings underscore the importance of implementing comprehensive AF management programs and warrant further evaluation through large-scale randomized controlled trials.Trial registrationClinicalTrials Identifier: NCT06337045.Supplementary InformationThe online version contains supplementary material available at 10.1186/s40814-025-01724-3.Keywords: Atrial fibrillation, Prevention, Rehabilitation, Complex intervention, Primary healthcare, Feasibility, Mixed-methods Key messages regarding feasibility What uncertainties existed regarding the feasibility?It was unclear whether the current set-up of cardiac prevention and rehabilitation in primary care settings in Denmark could effectively deliver a multidimensional prevention and rehabilitation intervention for individuals with atrial fibrillation. What uncertainties existed regarding the feasibility? It was unclear whether the current set-up of cardiac prevention and rehabilitation in primary care settings in Denmark could effectively deliver a multidimensional prevention and rehabilitation intervention for individuals with atrial fibrillation. What are the key feasibility findings?Conducting the intervention in a primary care setting yielded positive outcomes and acceptability of the intervention, demonstrating its feasibility in such settings. What are the key feasibility findings? Conducting the intervention in a primary care setting yielded positive outcomes and acceptability of the intervention, demonstrating its feasibility in such settings. What are the implications of the feasibility findings for the design of the main study?The findings of this feasibility study can form the basis for a large, randomised controlled trial to test, whether these findings can be replicated. What are the implications of the feasibility findings for the design of the main study? The findings of this feasibility study can form the basis for a large, randomised controlled trial to test, whether these findings can be replicated. Background The prevalence of atrial fibrillation (AF) is continuously increasing, and the condition poses a significant challenge globally [1, 2]. With a prevalence of 2–4% in adults and a 1 in 3-lifetime risk in Europe, the incidence and prevalence of AF have increased over the past two decades, mainly due to an increasing life expectancy and burden of risk factors and comorbidities [1]. This trend emphasizes the urgent need for effective management strategies. AF is associated with mental distress, reduced health-related quality of life (HRQoL), and a high rate of hospital readmissions, contributing to substantial healthcare costs; approximately 30% have at least one annual hospital readmission, and 10% have two or more [1, 3]. Research suggests that patient education [4–11] and physical exercise [12–14] might improve HRQoL and reduce symptom burden and hospital readmissions in individuals with AF. These interventions promoted self-efficacy, reduced insecurities and validated emotional responses in relation to living with AF [15]. In Denmark and other developed countries, the responsibility for preventive and rehabilitative interventions has moved across sectors from hospital level to a primary care setting [16, 17]. Although national and international guidelines recommend systematic referral to rehabilitation, including identification and management of risk factors and concomitant diseases, patient education and physical exercise [1, 18, 19], the integration of AF management into primary care provided rehabilitation lacks further qualification and evidence. To compare, cardiac rehabilitation (CR) for individuals with ischemic heart disease (IHD) and heart failure (HF) is well-established and based on a complex (i.e. multidimensional) intervention, including physical exercise, patient education, psychosocial healthcare, dietary intervention, smoking cessation and other risk factor management [20]. These interventions might also be relevant in AF. Patient-centered, integrated care has gained traction internationally, emphasizing the importance of a comprehensive approach to prevention and rehabilitation interventions in AF across sectors and healthcare providers [1, 21–24]. Integrated care follows a holistic and comprehensive approach, focusing on medical treatment as well as non-pharmacological interventions such as physical exercise, patient engagement including patient education, psychosocial management and lifestyle interventions. Also, Mind–body interventions such as yoga have shown promise in improving mental health, HRQoL and blood pressure in individuals with AF, and yoga was experienced as a usable tool to handle emotions and symptoms related to AF episodes [25–29]. Despite the existing evidence, there is still a need for comprehensive evaluation of these interventions, using mixed-methods approaches to provide a broader understanding of their impact. Moreover, it is unclear whether the current set-up of CR and prevention in primary care settings in Denmark can effectively deliver integrated, complex prevention and rehabilitation interventions for individuals with AF. Aim and objectives Therefore, the primary aim of this study was to investigate the feasibility of delivering such an intervention for individuals with AF in a primary Healthcare Center. Further objectives of the study were: To explore the participant’s experiences and acceptability of participating in the intervention. To explore the participant’s experiences and acceptability of participating in the intervention. To explore the participant’s needs and preferences for the intervention. To explore the participant’s needs and preferences for the intervention. To measure changes in HRQoL, anxiety and depression. To measure changes in HRQoL, anxiety and depression. To explore the feasibility of collecting patient-reported outcome measures as part of the intervention. To explore the feasibility of collecting patient-reported outcome measures as part of the intervention. Methods Design This study was designed as a prospective, non-randomized, single-arm, intervention, feasibility study. A convergent mixed-methods design was used, i.e. the quantitative and qualitative data were collected within the same timeframe to enable a more comprehensive understanding of the study results and feasibility [30]. Registration and reporting The study was retrospectively registered at clinicaltrials.gov (NCT06337045). The reporting was guided by the Consolidated Standards of Reporting Trials (CONSORT) extension to pilot and feasibility trials [31] (see checklist in additional file 1). Certain non-applicable areas of the CONSORT guideline were left out as according to Lancaster and Thabane [32]. Feasibility outcomes According to the CONSORT extension, a feasibility study aimed at informing a future randomized controlled trial (RCT) assesses whether a future trial can be done, should be done, and, if so, how it should be done [31]. The success of this study is determined by its ability to demonstrate feasibility within a primary care setting. Therefore, the primary feasibility outcomes of this study are to investigate the acceptability among participants and to assess whether there is a positive impact on key outcomes such as HRQoL, anxiety and depression. Our progression criteria related to the primary feasibility outcomes included stopping the study if intermediate analyses indicated increased anxiety, depression or decreased HRQoL due to the intervention, or if participants expressed significant worry or non-acceptability towards participation. In such cases, healthcare professionals (HCPs) working the primary Healthcare Center where the interventions were carried out would address the individual concerns by talking to the participant. If multiple such observation would occur, the study group agreed to convene to discuss the necessary actions, potentially including the termination of the study. Participants Participants were consecutively included from February to November, 2019 (i.e. ongoing enrollment), at a university hospital outpatient clinic during a consultation with an AF specialist nurse at a regular follow-up appointment. Patients were eligible for participation if they were diagnosed with AF at the cardiology department at the hospital and had symptomatic AF rated by the modified European Heart Rhythm Association (mEHRA) score [33]. An mEHRA-score ≥ 2B was the main inclusion criterion, i.e. participants were moderately affected (normal daily activity not affected but troubled by the symptoms), severely affected (normal daily activity affected) or disabled (normal daily activity discontinued) by AF. Besides eligibility based on the mEHRA-score, at least one risk factor (see Table 1) had to be present. The risk factors were determined by the AF nurse during the consultation based on data from patient records, patient-reported outcome measures (PROMs) and patient interviewing. Exclusion criteria were ongoing treatment adjustments due to HF, ongoing CR, or if participation was not possible due to dementia, mental illness, language barrier or comorbidity. Eligible participants gave informed consent for participation and were then referred to a complex prevention and rehabilitation intervention at a primary Healthcare Center. Figure 1 shows the flow of study and data collection timepoints. Table 1Inclusion criteria risk factors and measurement scale Risk factorsMeasured asHypertensionTreatment with > 1 drug, diagnosis verified by home-blood pressure measurement, 24-h ambulatory blood pressure measurement, or consultation blood pressure measurement > 160/90 mmHgDiabetes mellitusHbA1c > 58 mmol/mol unless higher values are accepted due to other aspectsObesityBody Mass Index (BMI) ≥ 30SmokingActive smokerAlcohol consumption > 7 units per week for women and > 14 units per week for men (according to The Danish Health Authority 2019 guidelines [34]), or motivation for reductionPhysical inactivityLess than the recommended 30 min per day [35]Sleep apneaKnown sleep apnea or STOP-bang Sleep Apnea Questionnaire score > 3 [36])Symptoms of anxiety or depressionAssessed using the self-reported PROM: Hospital Anxiety and Depression Scale (HADS) ≥ 8 [37]Mental vulnerabilityDefined as self-evaluated burdened life with high perceived stress, small social network, and/or low educational level (this item was inspired by references [38–40], which defines mental vulnerability as an individual with many psychosomatic symptoms and who finds it difficult to socialize Inclusion criteria risk factors and measurement scale Study flow and data collection time points Abbreviations: AF, Atrial Fibrillation, HCP, Healthcare Professional, PROMs, Patient Reported Outcome Measures. Box colours: blue: at the hospital; green: at the Primary Healthcare Center; orange: data collection time points The danish healthcare system The Danish healthcare system works across three levels: the state (national level), the regions (regional level) and the primary care settings administered by local goverments (local levels). The state sets the rules and oversees healthcare in general. The regions manage hospitals, including emergency care and psychiatry, as well as healthcare provided by general practitioners (GPs) and specialists in private practice. The local goverments handle primary healthcare and social services [41]. The local goverments offer comprehensive rehabilitation programmes for all citizens in need based in primary healthcare centres and includes: physical and mental training, patient education and other measures for restoring both functional abilities and the general ability to selfcare [41]. Setting This study took place at a primary Healthcare Center located in the region of Southern Denmark. The primary Healthcare Center is working with prevention, health promotion and rehabilitation for the following chronic diseases: heart disease, cancer, diabetes type-2 or chronic obstructive pulmonary disease. The team of HCPs working at the primary Healthcare Center is multidisciplinary, consisting of physiotherapists, occupational therapists, nurses and dieticians. Based on needs assessment and motivational interviewing during a consultation between the individual and a HCP, a complex intervention is offered at the primary Healthcare Center [42]. This approach was also used in this study. The complex intervention normally includes physical exercise, smoking cessation, alcohol consumption counselling, dietary interventions and/or patient education based on individual needs [43], i.e. these are Danish-national recommendations. Complex intervention The Medical Research Councils (MRC) guidelines for complex interventions [44] were the overarching framework of the study. The complex intervention (afterwards called “the intervention”) included the national recommended interventions described above, although with the patient education targeted at AF. Furthermore, we included Medicinal yoga (MediYoga) due to the promising evidence of mind–body interventions and group-based psychosocial support due to the association between AF and mental distress. The interventions are in accordance with current international and national guidelines [1, 18, 45]. All participants started and ended their intervention with a consultation with a HCP (nurse or occupational therapist) in the primary Healthcare Center. The elements and duration of the intervention are described in Fig. 2. All activities were optional and based on the individual needs assessment done in the starting consultation with the HCP, and the participants’ motivation and preferences towards the intervention components. Thus, the total time spent as part of the intervention varied for each participant and all participants entered into their individual collection of interventions once referral procedure was done. Furthermore, if the participant found it relevant and important to their treatment course, significant others were allowed to participate in the patient education, group-based psychosocial support interventions and in the individual consultations with the HCP. Only some participants chose this option and we did not collect any data or had separate questions for significant others. The complex intervention in the primary Healthcare Center Abbreviations: HCPs, Healthcare Professionals, AF, Atrial Fibrillation, MediYoga, Medicinal Yoga Data collection Sample size As the primary objective was investigating feasibility, no formal sample size calculation was performed. According to Billingham et al., the median sample size per arm reported in feasibility studies are 36 (IQR: 25–50) [46]. As our study was single-armed, we sought to reach a minimum of 36 participants. Related to the primary feasibility outcomes, we assessed that a sample size of minimum 36 participants would be sufficient to assess impact on the key outcomes and acceptability among participants. Quantitative Baseline characteristics of the participants were collected from hospital patient records. Two PROMs were used and collected at three time points: T0 (hospital consultation with the AF nurse), T1 (the end of the intervention, in the primary Healthcare Center) and T2 (two-month follow-up) (Fig. 1). The participants also responded to four evaluative questions collected at T1. An overview of outcomes and measurement instruments is shown in Table 2. PROMs were administered by SurveyXact [47]. Paper-based versions were used at T0 and T1 if participants had not completed the online version before the consultations and subsequently entered into SurveyXact. At T2, the participants were at home, and non-respondents were phoned. Table 2Overview of outcomes and measurement instrumentsOutcomeInstrumentItems and scalesMeasured at timepointHealth-related Quality of LifeHeartQoL14 items, 3 scales (global, physical and emotional). Measured on a 4-point Likert scale from 0–3, where a higher score represents better HRQoLT0, T1, T2Symptoms of Anxiety and DepressionThe Hospital Anxiety and Depression Scale (HADS)14 items, 2 scales (anxiety and depression). Measured on a 4-point Likert scale from 0–3, where 0 reflects the positive extreme (no symptoms), and 3 reflects the negative extreme of the scale (symptoms)T0, T1, T2Four evaluative questionsSelf-made instrument for participant evaluation of the total intervention:1) Did the HCPs listen to you?2) Did the HCPs prioritize what mattered most to you?3) Did you experience a coherence in the intervention?4) Did you experience the intervention as relevant to your situation?Rated on a 5-point Likert-scale with the values “Very much”, “Much”, “Maybe”, “Little”, and “Very little”T1 Overview of outcomes and measurement instruments Self-made instrument for participant evaluation of the total intervention: 1) Did the HCPs listen to you? 2) Did the HCPs prioritize what mattered most to you? 3) Did you experience a coherence in the intervention? 4) Did you experience the intervention as relevant to your situation? Measures The PROMs were the HeartQoL questionnaire measuring HRQoL [48, 49] and the Hospital Anxiety and Depression scale (HADS) measuring symptoms of anxiety and depression [37, 49]. HeartQoL contains three scales, a global overall scale and a physical- and emotional sub-scale. It has a 14-item self-assessment scale enabling the respondents to recall how much their heart problems have bothered them during the past four weeks on a 4-point Likert scale from 0–3, where a higher score represents better HRQoL. The global score reflects all items and is calculated by dividing the sum of the scored items by the number of scored items. The same principle applies to the physical and emotional subscales, which hold 10 items and 4 items, respectively [48, 49]. HADS is a 14-item self-assessment scale, 7 items for each domain. All items are rated on a 4-point Likert scale, where 0 reflects the positive extreme and 3 reflects the negative extreme of the scale. For each domain, a score ranging from 0 to 21 can be obtained and categorized as “normal” (0–7 points), “possible presence of a disorder” (8–10 points), and “probable presence of a disorder” (11–21 points) [49]. Lastly, four self-made questions for participant evaluation of the total intervention were constructed and applied at T1 (Table 2). All four questions were rated on a 5-point Likert-scale with the values “Very much”, “Much”, “Maybe”, “Little”, and “Very little”. Qualitative Focus group interviews were conducted twice among a convenience sample just before T1 (Fig. 1). A semi-structured interview guide covering participant experiences and acceptability, rehabilitation needs and intervention preferences was used. In total, 11 participants were invited to the two interviews randomly selected among participants who had almost finished the intervention at the time of the interview. Thereby minimizing their recall period as well as gaining a detailed picture of their experiences with the total intervention. Interviews were recorded and transcribed. Names were changed to ensure anonymization. One of the authors (TS), a nurse with interviewing experience, conducted the focus group interviews. The author (CME) took part in the interviews as observer of the group dynamics and ensured everybody could contribute with their experiences. Furthermore, field notes from HCPs collecting the data were used to evaluate the feasibility of collecting PROMs. Field notes were written during the intervention and collected at the end of the study. Data analysis Quantitative Descriptive statistics (median, range, frequencies and percentages) were used to describe participant characteristics, needs and choices of interventions as well as levels of HRQoL and symptoms of anxiety and depression. The results of the four evaluative questions (see Table 2) were presented as frequencies. Qualitative Based on the research questions, we conducted an inductive, semantic thematic analysis guided by Braun and Clarke’s [50] six stages of thematic analysis. Both the focus group interview data and the field notes were integrated in the analysis. As the analysis was based on the research questions, codes were defined a priori as “Experiences of participating and acceptability”, “Needs and preferences of interventions”, “Changes in HRQoL, anxiety and depression” and “Collecting PROMs”. Based on these initial codes, the interviews and field notes were inductively and semantically analyzed, as the analysis was driven by the data content and explored meaning at the explicit level based on what was actually said [50]. Relevant data extracts were pooled under these codes and finally into themes. Two authors (CME and MKP), independently of each other, analyzed the qualitative data based on the codes and reflected on themes. Three of the authors (CME, MKP and SSR) met and discussed the themes. NVivo 14 [51] was used for analysis of the qualitative data. A Table presented in additional file 2 describes the entire analysis process. Mixed-methods analysis The mixed-methods analysis involved merging the quantitative and qualitative data according to the secondary objectives of the study into side-by-side joint displays (visual presentation) [52], followed by a thorough discussion of the results. A quantitative and qualitative summary was provided along with mixed-methods inferences (meta-inferences) to evaluate the alignment between the findings. The alignment was evaluated as either confirmation (findings support each other), expansion (findings provided further insights) or discordance (findings contradicted each other) [30, 52]. Results An adjusted CONSORT flowchart shows the flow of the study participants (Fig. 3). In total, 50 individuals agreed to participate, and 43 participants (86%) completed the intervention. Thus, the required minimum sample size was reached. Table 3 presents the baseline characteristics of the participants who completed the intervention. The median age was 66 years (range from 52–81 years) and 53% were female. Most (49%) were retired from work and living with partner (58%). Flowchart Abbreviations: HF, Heart Failure, CR, Cardiac Rehabilitation. *These were six males and one female. Age from 53 to 76 years. Two were still working and one seeking employment, four retired. Six of them were overweight/obese; one smoked and over-consumed alcohol. Their T0 HeartQoL global was 2.64, HADS-A was 2, HADS-D was 4. Flowchart based on the CONSORT extension to pilot and feasibility trials Table 3Participant characteristics at baselineN = 43VariableLevelMedian (range)Number (percentage)AgeContinuous66 (52–81)SexMale20 (47)Female23 (53)Inclusion criteriamEHRA1-scoreIIB13 (30)III30 (70)Risk factorsHypertension2Yes18 (42)Diabetes3Yes3 (7)ObesityBMI4 ≥ 3016 (37)BMI, continuous28.1 (19.5–45)SmokingYes7 (16)Alcohol overconsumption5Male overconsumption1 (2)Female overconsumption3 (7)Physical inactivity6Less than recommendation36 (84)Sleep apnoea7STOP-bang score > 325 (58)STOP-bang score, continuous3 (0–7)Anxiety or depression8HADS-A or HADS-D ≥ 819 (44)HADS-A, continuous7 (0–15)HADS-D, continuous4 (0–13)Mental vulnerability9Yes22 (51)More than one risk factorYes37 (86)Type of AFParoxysmal24 (56)Persistent12 (28)Permanent4 (9)Atrial flutter3 (7)Time since diagnosisLess than one year16 (37)More than one year27 (63)Experience of relapsesYes29 (67)CHA2DS2-VASc Score10Low risk, male/female1 (2) / 7 (16)Moderate risk, male/female9 (21) / 8 (19)High risk, male/female10 (23) / 8 (19)Civil statusLiving with partner25 (58)Living alone18 (42)Occupational statusWorking11 (26)Reduced hours working (flexible job)3 (7)Sick leave4 (9)Retired21 (49)Early retirement4 (9)1mEHRA: modified European Heart Rhythm Association-score2Hypertension (treatment with > 1 medicament, or diagnosis verified by home-blood pressure or ambulatory blood pressure in 24 h or if blood pressure is measured in the AF-clinic > 160/90)3Diabetes mellitus (HbA1c > 58 mmol/mol unless higher values are accepted due to other aspects)4BMI: Body Mass Index5Alcohol consumption (> 7/14 units per week for women/men [34], or motivation for reduction)6Physical inactivity (less than the recommended 30 min per day [35])7Known sleep apnea or STOP-bang Sleep Apnea Questionnaire score > 3 [36])8Anxiety or depression (the Hospital Anxiety and Depression Scale (HADS) > 7 [37])9Mental vulnerability (inspired by [38–40]), locally translated to self-evaluated strain in life based on low educational level, small social network, high perceived stress, etc.10CHA2DS2-VASc Score [1]: Low risk is a score of 0 in men and 1 in women. Moderate risk is a score of 1 in men and 2 in women, oral anticoagulation therapy should be considered. High risk is a score of 2 or higher in men, and 3 or higher in women, oral anticoagulation therapy is indicated Participant characteristics at baseline 1mEHRA: modified European Heart Rhythm Association-score 2Hypertension (treatment with > 1 medicament, or diagnosis verified by home-blood pressure or ambulatory blood pressure in 24 h or if blood pressure is measured in the AF-clinic > 160/90) 3Diabetes mellitus (HbA1c > 58 mmol/mol unless higher values are accepted due to other aspects) 4BMI: Body Mass Index 5Alcohol consumption (> 7/14 units per week for women/men [34], or motivation for reduction) 6Physical inactivity (less than the recommended 30 min per day [35]) 7Known sleep apnea or STOP-bang Sleep Apnea Questionnaire score > 3 [36]) 8Anxiety or depression (the Hospital Anxiety and Depression Scale (HADS) > 7 [37]) 9Mental vulnerability (inspired by [38–40]), locally translated to self-evaluated strain in life based on low educational level, small social network, high perceived stress, etc. 10CHA2DS2-VASc Score [1]: Low risk is a score of 0 in men and 1 in women. Moderate risk is a score of 1 in men and 2 in women, oral anticoagulation therapy should be considered. High risk is a score of 2 or higher in men, and 3 or higher in women, oral anticoagulation therapy is indicated The results of PROMs, based on median score changes, show that HRQoL increased from baseline (T0: 1.93) to the end of intervention (T1: 2.36) and maintained this level two months after ending the intervention (T2: 2.43). Symptoms of anxiety and depression decreased from baseline (T0, HADS-A: 7 and HADS-D: 4) to the end of intervention (T1, HADS-A: 3 and HADS-D: 1) and maintained this level two months after ending the intervention (T2, HADS-A: 4 and HADS-D: 2). For the possible presence of symptoms of anxiety and depression indicated by a HADS score ≥ 8, results show that the percentage decreased accordingly. These results can be seen in the table presented in Figure 7. Joint display regarding changes in HRQoL, anxiety and depression Qualitative results The eleven informants in the two focus group interviews were six females and five males. They were aged 52–79 years. A Table in additional file 3 presents age and sex information for each informant ID number. Figure 4 presents the codes and themes of the thematic analysis. In the end, nine themes were agreed upon and the headings of the themes were: Being understood and supported; Including family members in rehabilitation; What comes next; Knowing what you need; Needs and preferences related to specific interventions; Gaining tools and knowledge from the intervention; Relation; Using data; and Academic systematism. Codes and themes of the thematic analysis Abbreviations: HRQoL, Health-related Quality of Life, PROMs, Patient Reported Outcome Measures. The four bobbles in the shaded area represents the codes and the white area represents the themes Mixed-methods results The side-by-side joint displays (Figures 5, 6, 7 and 8) showed that the quantitative and qualitative results mutually expanded the understanding of the findings. The joint display regarding “experiences of participating and acceptability” related to the primary care complex intervention showed that the participants felt listened to and supported by HCPs and peers, which were very valuable and led to an experience of coherence (Fig. 5). The joint display regarding the “needs and preferences of the interventions” showed that it was difficult for the participants to know what they needed when they had just been diagnosed with a heart disease, and the most important thing was the availability of a prevention and rehabilitation offer (Fig. 6). The joint display regarding “changes in HRQoL, anxiety and depression” showed the participants’ decreased symptoms of anxiety and depression and increased HRQoL, and the participants experienced that gaining tools and knowledge from the interventions originated these effects (Fig. 7). The joint display regarding the feasibility of “collecting PROMs” pointed to some important aspects of a successful data collection, such as establishing and maintaining a trusting relationship with the participants and using PROMs for purposes other than statistics like consultation tools and tools for case discussion across sectors (Fig. 8). Joint display regarding the experience of participating in the intervention and acceptability Joint display regarding needs and preferences of the interventions Joint display regarding the feasibility of collecting PROMs Harms There were no adverse events reported (deaths, hospital readmissions, unintended consequences or harms) due to the intervention during the study. This evaluation was based on “nothing mentioned” by HCPs in cooperative project meetings between the HCPs in the primary Healthcare Center and HCPs from the hospital department as well as no participants reported any adverse events or lack of acceptability during the study. Furthermore, the results of the PROMs did not indicate that the participants were affected negatively by participating. Thus, no observations were made that would terminate the study. Discussion The findings of this study provide valuable insights into the feasibility, acceptability and potential outcomes of a complex prevention and rehabilitation intervention for individuals with AF provided in a primary Healthcare Center. Conducting the intervention in a primary care setting yielded positive results, demonstrating its feasibility, acceptability and viability in such settings. The participants improved outcomes in terms of HRQoL, anxiety and depression and the intervention was well received as participants felt supported and highlighted the importance of the intervention. Similar findings were shown in another feasibility study from a real-world, English set-up testing an exercise-based cardiac rehabilitation intervention with education sessions. The study found lower anxiety and depression and improved six-minute walk test [53]. Our mixed-methods analysis provided an expanded understanding of the findings revealing the significance of support from HCPs and peers as well as increased knowledge of AF and tools to tackle AF, which contributed to the perceived effectiveness of the intervention. Additionally, participants' highlighted relying on individual consultations with HCPs to determine relevant interventions, emphasizing the importance of a tailored, comprehensive care offer. Furthermore, the feasibility of collecting PROMs during the intervention was demonstrated, with participants willing to respond and valuing their use beyond statistical purposes. The study adhered to the guidelines recommending multi-dimensional, integrated, tailored care [1, 19] across sectors and disciplines [1, 18]. Thus, in this study, participants received evidence-based care tailored to their individual needs. The intervention specifically targeted the management of risk factors and lifestyle modifications, patient education including medication adherence, and psychosocial management, as according to the integrated care model [1]. The medical treatment (i.e. stroke prevention, symptom control and management of comorbidities) was managed at the hospital. The positive outcomes observed in this study provide additional support for the efficacy of integrated care with guideline-concordant interventions, which is also shown in other studies (however, mostly focused on physical exercise and patient education) [5, 8, 9, 54–56]. Furthermore, the results of a meta-analysis showed a 49% and 42% decrease in all-cause mortality and cardiovascular hospitalizations, respectively, with integrated care [22]. Implementation of integrated care requires a change in the approach to patient care, which involves key stakeholders as well. This can be further discussed by looking at the framework for complex interventions [44]. This study went through the development of the intervention and investigation of its feasibility always considering the core elements of the framework like context, key stakeholders and economic considerations. To complete the framework, the intervention still needs large-scale evaluation and implementation. Based on this feasibility study, we can already consider core elements related to implementation. The Danish context in which the integrated complex intervention was provided is dynamic and multi-dimensional. Key dimensions include physical, spatial, organizational, social, cultural, political and economic features in the healthcare system [44]. In many European countries, key stakeholders representing the context are politicians who allocate funds to specific objectives based on a political agreement. This applies to both the state, region, and local levels [41]. In the local governments, the politicians’ need to be convinced that the limited funds available should also be allocated to support an integrated, complex intervention set-up in AF. Thus, making the politicians confident with the efficacy of the intervention to get it on the political agenda is key to implementation. A good and valid argument could be cost savings related to AF. A study focusing on costs in the ‘Atrial fibrillation Better Care’ (ABC) pathway showed a cost saving of 542 dollars per 1-patient-year for ‘C’ vs. ‘non-C’ in a non-fatal cardiac event or ischemic stroke [57]. ‘C’ means the identification and management of concomitant diseases, risk factors and unhealthy lifestyle which prevents and reduces AF burden and symptom severity [1]. In Denmark, the local governments co-finance the hospitals [41], and it would be a prioritization to lower the costs of hospital readmissions. The HCP hours were not calculated in this study, but it would be important to consider the costs of educating the HCP on this specific patient group and the HCP hours to deliver the intervention. Methodological considerations The study's generalizability should be considered in light of the recruitment strategy and intervention design. Participant selection from the hospital outpatient clinic based on AF symptoms and risk factors introduces selection bias, limiting applicability to the broader AF population. Also, it is well-known in rehabilitation research that it tends to attract participants who may not be socially vulnerable, in low socio-economic status, and/or with high comorbidity, etc., potentially skewing the findings [58]. However, the intervention's real-world resemblance in primary care settings suggests potential applicability to similar settings. Strengths of the study include adherence to established methodological guidelines such as MRC guidelines for complex interventions [44] and CONSORT extension to pilot and feasibility trials [31], ensuring the methodological robustness. The convergent mixed-methods design allowed for a nuanced understanding of the feasibility, acceptability and potential effectiveness. Merging the quantitative and qualitative data into joint displays revealed confirming and expanding meta-inferences and led to a better understanding of the study results [30, 52]. Allthough it can be stated that our thematic analysis approach was initially deductive (through the lens of the objectives), we believe it was a strength that we continued in an inductive, semantic approach, driven by the data content and exploring meaning by what was actually said. This opened our analysis to focusing on what was at stake and important to the participants. Thus, the analysis provided a detailed examination of the qualitative data [50]. We reached the minimum inclusion target of at least 36 participants [46], and together with the use of validated PROMs and achieving a high response rate, this enhanced the reliability of the outcomes. The enrollment was ongoing, which we believe a strength of the study, as it replicated the real-world scenario. Some limitations must be acknowledged. First, in many mixed-methods studies, data collection is designed to explore the same themes. For example, an interview guide may elaborate on the items in PROMs, which helps the process of integrating results directly with each other. However, this study did not follow that approach, as it was not the purpose, and the interview guide was designed according to the secondary objectives. Though limiting direct integration of results [30]. Second, though reaching 43 participants, the small sample size and non-normal distribution of data precluded formal statistical analysis of intervention effects or using measures like minimal important difference [59], thus warranting descriptive analysis. Third, absence of a control group means that we could not out-rule that the observed effect could have occurred by itself over time. Participants in the interviews highlighted the value of being with peers and we could not out-rule a peer-to-peer effect either. Another study showed increased HRQoL and decreased anxiety and depression following a peer-mentor intervention among individuals with myocardial infarction [60], indicating potential peer-to-peer effect. Fourth, inability to evaluate individual intervention effects are a limitation to the study. Had many more participants been included in the study, it would have been relevant to conduct multifactorial analysis to determine if a specific intervention was more effective than others [61]. Fifth, a small proportion of missing data was seen at T2 as we had nine individuals (21%) who did not respond to PROMs at T2, and this warrants consideration for the impact on the overall findings. Applying their scores from T1 at T2 (assuming no further increase or decrease) did not change the median results on anxiety and depression, and the results on HRQoL decreased a bit but were almost the same compared to the table presented in Figure 7 at T1 (see additional file 4). Thus, the impact of missing data is considered limited. Implications for clinical practice The findings of this study showed that it was feasible to conduct a complex prevention and rehabilitation intervention for individuals with AF in a primary Healthcare Center. Thus, clinical practice can draw inspiration from this study to implement similar interventions for individuals with AF working closely together in primary and secondary care. HCPs can also draw inspiration from the insights gained from the participants’ experiences and preferences to tailor care to individual needs as recommended in international guidelines [1]. A concrete example is that of the participants finding it difficult to know what they need by themselves, and therefore rely on the guidance of HCPs. Moreover, collecting PROMs underlined the importance of incorporating individual perspectives into AF management in clinical practice and is also recommented by experts in the field [62], this study gives a concrete example of how to do this in practice. Future research Future research should prioritize large, multicenter RCTs based on a sample size calculation, testing a complex intervention design to confirm and expand the promising results of this study. Adopting a complex intervention approach tailored to the multifaceted nature of AF is essential, incorporating various components such as psychosocial support, mind–body interventions and lifestyle modifications. Utilizing mixed-methods and multifactorial designs can provide a more comprehensive understanding of intervention impact, accounting for the diverse factors influencing outcomes in individuals with AF. By further evaluating the intervention's efficacy and scalability, such RCT can provide valuable evidence to inform the development and implementation of comprehensive AF prevention and rehabilitation programs. These strategies will further enhance the validity and generalizability of study findings, potentially impacting management approaches for individuals with AF. Conclusion In conclusion, this study found it feasible to deliver a complex prevention and rehabilitation intervention for individuals with AF in a primary care setting. Through the exploration of participant experiences, needs and preferences, as well as the assessment of changes in HRQoL, anxiety and depression, valuable insights were gained into the potential effectiveness and acceptability of the intervention. Moreover, the feasibility of collecting PROMs as part of the intervention was demonstrated. These findings lay a solid foundation for implementing a future full-scale RCT. Supplementary Information Additional file 1. Additional file 2. Additional file 3. Additional file 4. Additional file 5. Acknowledgements The authors would like to acknowledge and thank all participants who took part in this study. Thanks to the nurses at the atrial fibrillation clinic at Svendborg Hospital, Odense University Hospital, Region of Southern Denmark, for including participants to the study. Thanks to the healthcare professionals working in the primary Healthcare Center in the local government Svendborg, Region of Southern Denmark, for conducting the intervention and collecting data. Thanks to the Danish Heart Association at Funen and their local committee in Svendborg for participating actively with advice and guidance, and for allocating representatives who participated and peer-mentored with the participant. Abbreviations Atrial fibrillation Better Care Atrial fibrillation Body Mass Index CHA2DS2-VASc Score for Atrial Fibrillation Stroke Risk The Consolidated Standards of Reporting Trials Cardiac rehabilitation European Heart Rhythm Association European Society of Cardiology General practitioners The Hospital Anxiety and Depression Scale Heart failure Healthcare professionals Health-related quality of life Ischemic heart disease Medicinal yoga Medical Research Councils Patient reported outcome measures Randomized controlled trial Authors’ contributions CME: Design, acquisition of data, data analysis, interpretation of data, responsible for drafting the manuscript. MKP: Data analysis, interpretation of data, drafted manuscript, revised manuscript. ADOZ: Conception, design, interpretation of data, revised manuscript. BB: Interpretation of data, revised manuscript. JFE: Conception, design, interpretation of data, revised manuscript. TS: Conception, design, acquisition of data, interpretation of data, revised manuscript. CT: Conception, design, interpretation of data, revised manuscript. JMH: Interpretation of data, revised manuscript. AB: Interpretation of data, revised manuscript. SSR: Conception, design, data analysis, interpretation of data, drafted manuscript, revised manuscript. All authors approved the final submitted manuscript and agree to be personally accounTable for their contributions and to the accuracy and integrity of the work. Funding Open access funding provided by University of Southern Denmark The primary Healthcare Center in cooperation with the Department of Cardiology at the local hospital (2018–2020), received funding from the Danish Health Authority to develop, trial and evaluate a complex intervention aimed at individuals with AF. This project is part of a PhD partially funded by a PhD faculty scholarship from the University of Southern Denmark, a PhD stipendium from the Region of Southern Denmark and running costs from the fund to support independent, strategic clinical research in the Region of Southern Denmark. Data availability The datasets analyzed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The study was retrospectively registered at clinicaltrials.gov (NCT06337045). The study conformed with the principles outlined in the Declaration of Helsinki. Participants have provided written informed consent to participate in the study and data collection. The Danish Health Authority approved the use of the data in this study. Data were collected during the year 2019. The study is registered at the Regional Committees on Health Research Ethics for Southern Denmark (20222000–24) as well as at the Region of Southern Denmark’s record of data processing activities (22/28369). Consent for publication Along with the consent to participate, the participants were informed about the publication of results and also consented to this. Competing interests The authors declare that they have no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Additional file 1. Additional file 2. Additional file 3. Additional file 4. Additional file 5. Data Availability Statement The datasets analyzed during the current study are available from the corresponding author on reasonable request.