The Atrial Fibrillation Better Care pathway for managing atrial fibrillation: a review

Abstract The 2020 European Society of Cardiology guidelines endorse the Atrial Fibrillation Better Care (ABC) pathway as a structured approach for the management of atrial fibrillation (AF), addressing three principal elements: ‘A’ – avoid stroke (with oral anticoagulation), ‘B’ – patient-focused better symptom management, and ‘C’ – cardiovascular and comorbidity risk factor reduction and management. This review summarizes the definitions used for the ABC criteria in different studies and the impact of adherence/non-adherence on clinical outcomes, from 12 studies on seven different cohorts. All studies consistently showed statistically significant reductions in the risk of stroke, myocardial infarction, and mortality among those with ABC pathway adherent treatment. The ABC pathway provides a simple decision-making framework to enable consistent equitable care from clinicians in primary and secondary/tertiary care. Further research examining the impact of ABC pathway implementation in prospective cohorts utilizing consistent inclusion criteria and definitions of ‘A’, ‘B’, and ‘C’ adherent care is warranted.


Introduction
Atrial fibrillation (AF) is associated with a five-fold increase in the risk of stroke 1 and a higher risk of cardiovascular and all-cause mortality. 2 Current European Society of Cardiology (ESC) guidelines on AF management advocate the use of oral anticoagulants (OACs) to reduce stroke risk in patients with a CHA 2 DS 2 -VASc risk score of > _1 for men and > _2 for women. 3 More recently, there has been a move towards recommending an integrated care approach to AF management. [3][4][5][6][7][8] Three studies examining integrated care for the management of AF [6][7][8] were analysed in a meta-analysis, which showed a significant reduction in the risk of both mortality and hospitalization 9 ; however, this systematic review showed inconsistency in the populations recruited and the care provided between the studies.
In 2017, the Atrial Fibrillation Better Care (ABC) pathway was proposed as an integrated, structured approach to AF management, 10 addressing three main components: 'A' refers to 'avoid stroke', by offering stroke prevention with appropriate OAC to patients with a CHA 2 DS 2 -VASc score of > _1 for men and > _2 for women. 1 . 'B' refers to 'better symptom management' and involves a patient and symptom-focused approach to decisions on managing heart rate or rhythm. 'C' refers to 'cardiovascular and comorbidity risk reduction', comprising the management of risk factors for other cardiovascular outcomes.

Results
The searches for this review returned 19 studies and after reviewing the titles and abstracts, 12 studies 12-23 were reviewed as full-text and included. Reasons for exclusion included: reviews (n = 2), guidelines (n = 2), ABC criteria not defined (n = 1), wrong population and no reference to ABC pathway (n = 1), and wrong outcomes (i.e. costs) (n = 1). The 12 included studies used data from seven different datasets. Three datasets were prospectively collected, 12,15-18 two were retrospective post hoc analyses of prospectively collected data 19,21,22 and two were registries or electronic health records. 13,14,20,23 Characteristics of the included studies are provided in Table 1. Studies used data from around the world: South Korea (n = 3), 13,14,23 China (n = 2), 17,18 the Middle East (n = 2), 15,16 Italy (n = 1), 19 Europe (n = 1), 20 the USA and Canada (n = 2), 21,22 and the Balkans (n = 1). 12 Sample sizes varied from 603 in the Gulf Survey of Atrial Fibrillation Events (SAFE) Registry 15 to over 260 000 in the Korea National Health Insurance Service database. 13 Age varied considerably between studies, ranging from 56.7 16 to 73.1 years. 19 Two studies had a difference of over 8 years in mean age between ABC adherent and non-ABC adherent patients. 13,16 The proportion of women included in each study ranged from 37.5% 14 to 52.2%. 15 The follow-up times of six of the studies were relatively short, at only 1-2 years. [15][16][17][18][19][20] Only the studies based on the Korean Nation Health Insurance Service database 13,14,23 and the AFFIRM trial 21,22 followed up patients for >2 years. There was no significant difference between the results of studies with longer and shorter follow-up. However, there was no indication that studies had tested that the risk reduction due to ABC adherence remained constant over time although they used models that assumed proportional hazards.
'A'-avoid stroke with oral anticoagulation All studies required OAC prescription for patients to be based on stroke risk identified with the CHA 2 DS 2 -VASc score. The definition of a high risk of stroke varied between studies. To meet the criteria for the 'A' component, one study considered OAC optional for patients with a CHA 2 DS 2 -VASc of 1 or 2 for men or women, 19 respectively, while others considered that OAC was required in these patients. 13,16,20,23 Five studies only included patients that had a CHA 2 DS 2 -VASc score > _1 or > _2 for men or women, respectively, meaning that all patients were eligible for OAC. 15,17,18,21,22 Each study defined OAC adherence using different criteria. For patients receiving warfarin or other vitamin K antagonists (VKAs), time in therapeutic range (TiTR) was utilized to indicate anticoagulation control by five papers. [17][18][19]21,22 For three studies, [17][18][19] the target TiTR was >65% and in two others 21,22 the target was >70%. TiTR was not always available; alternatively, prescription days coverage >80% 13,14,23 was used.

'B'-better symptom management
Seven studies defined adherence to the 'B' criterion as symptom levels classified as European Heart Rhythm Association (EHRA) classes I-II. 12,[15][16][17][18][19][20] Studies using the AFFIRM trial data allowed < _2 symptoms from their own list. 21,22 The studies based on the Korea National Health Insurance Service database did not have data on symptoms, therefore the authors used the criteria of <5 outpatient visits per year as a proxy. 13,14,23 'C'-cardiovascular and co-morbidity management Each study considered a different set of conditions when defining the 'C' criteria as shown in Table 2 although it was defined in multiple ways. Nine studies required blood pressure (BP) to be controlled at <140/90 mm Hg 12-16,20-23 although other cut-offs (e.g. 160/90 19 or 140/85 17,18 ) were used. [17][18][19] Two studies looked for active treatment of hypertension with pharmacological treatment rather than BP control. 12,19 Each study looked at a different selection of other conditions such as diabetes, [12][13][14]16,19,20,23 heart failure, 12-23 peripheral artery disease, [13][14][15][16][20][21][22][23] and coronary artery disease 12,15,16,[20][21][22] ; these were considered based on drugs used for prevention and/or treatment. Body mass index with a cutoff of 30 kg/m 2 was considered for obesity in three studies. 13,14,23 There was a wide-range in the proportion of participants assessed as ABC adherent in the included studies (7.0-43.8%), 12,21,22 as shown in Table 3. Mean age varied among studies depending on the inclusion criteria. In three studies, those who were ABC adherent were over 10 years younger 12-14 than those who were not ABC adherent; conversely in another study, ABC-adherent patients were over 8 years older. 16 In four studies a lower proportion of ABC adherent patients were women, 14,[20][21][22] while in two studies a higher proportion were women. 12,19 Hypertension was more prevalent in ABC non-adherent patients, although this was dependent on definitions. Table 4 presents the outcomes in ABC-adherent vs. non-ABC adherent patients within each study. Each study adjusted for a different set of potential confounders, although age, sex, and diabetes status were adjusted for in eight of the studies. 13,14,17,18,20-23 Due to different data availability, both Cox proportional hazards models and logistic regression were used to estimate the effect of ABC adherence on clinical outcomes. Hazard ratios (HRs) and odds ratios varied due to differing definitions but consistently reported that ABC pathway adherent care was beneficial for lowering mortality [  14,20,21 There was also a risk reduction for cardiovascular mortality 20 and composite outcomes. 14,20-22 There was a consistent dose-  response effect with more ABC-adherent criteria fulfilled translating into a lower risk for all outcomes. 14,20-22

Discussion
All nine studies that examined the risk of adverse outcomes among patients adherent to the ABC pathway reported a significant risk reduction of adverse events, with only one study showing a non-significant result for major bleeding. 13 The risks of stroke, mortality, myocardial infarction, hospitalization, and composites of these outcomes have all been shown to be lower in patient's adherent to the ABC pathway. None of the studies suggested that there was any negative effect of being adherent to the ABC pathway. The significant positive effect of ABC pathway adherence was robust amongst the different datasets. However, there was a relatively large variation in the strength of the risk reduction (e.g. HRs ranged from 0.35 to 0.93 for mortality), reflecting the differences between the datasets, and criteria used to denote A, B, and C adherence which may result in differences in the degree of risk reduction. Several factors could be driving variation, for example, some of the studies only included patients with other stroke risk   Pastori (2019)      factors (e.g. older age or diabetes) and some studies used more robust definitions for ABC adherence. Seven of the included studies conducted a retrospective analysis of pre-existing datasets. 13,14,[19][20][21][22][23] The various retrospective analyses led to variation between the studies examined within this review including differences in the inclusion/exclusion criteria, definitions of ABC-adherence employed and study design. Lack of appropriate data, such as TiTR, AF symptoms, and treatment data for each of the criteria of the ABC pathway included, led to some studies using less comprehensive definitions 13,14,23 than others. 12,[15][16][17][18][19][20][21][22] Care is needed when defining the 'A', 'B', and 'C' criteria to be used in retrospective studies as there is also the potential for healthier patients to be selected rather than just those who have had ABC adherent management. Not all criteria can be modified quickly after AF diagnosis and some require patient involvement, such as adherence to prescriptions, increasing TiTR, and reducing risk factors such as obesity.
All studies only examined if the patient's care was adherent to the ABC pathway at baseline. However, risk factors have the potential to change over time, 24 especially in patients that were newly diagnosed with AF at baseline. In studies with longer follow-up, changes from baseline are more likely. There was a large variation in follow-up length in the studies in this review, although all but two datasets had follow-up < _2 years. 13,14,[21][22][23] Although all studies adjusted for the patient's age when analysing the risk of adverse outcomes in patients adherent and non-adherent to the ABC pathway, only one stratified the results by different age groups. 23 The results of this study suggested that there may be a greater risk reduction in older patients, but the study lacked power for this analysis.
Wagner et al. 25 first purported the idea of integrated care for chronic diseases in 1996. The key to integrated care is engaging the patient in the decision-making process and management of their condition. Also crucial is involving a multidisciplinary team from specialists to carers in the success of AF management. These strategies aim to improve treatment adherence, reduce perceived treatment burden and provide better outcomes for the patient.
While some of the individual components that comprise the ABC pathway have previously been included in guidelines, 26 the ABC pathway has recently been incorporated into the 2020 ESC guidelines for the management of AF, 3 bringing these together in an easy to follow structure. This review adds to the evidence supporting the inclusion of the ABC pathway in AF guidelines and implementation in practice to improve patient outcomes. The heterogeneity of the retrospective cohorts and the ABC pathway assessments based on available data and outcomes are intrinsic to the particular studies; this could be avoided by prospective studies. The mAFA-II cluster randomized trial compared usual care against app-based mobile health (mHealth) intervention based on the ABC pathway 18 and showed a risk reduction for those using the app-based care of 61% for a composite outcome of stroke/ thromboembolism, all-cause mortality, and re-hospitalization and a risk reduction of 68% for re-hospitalization.
The long-term mAFA-II cohort showed high adherence and persistence of use, and maintenance of improved clinical outcomes with ABC pathway adherent management. 17

Strengths and limitations
This review has summarized all available studies that have examined the impact of ABC adherent vs. non-ABC adherent treatment in AF patients, showing a consistent clinically significant reduction in the risk of adverse outcomes for patients whose treatment is adherent to the ABC pathway. However, variation between the studies included in this review raises questions over the precise magnitude of the benefit of adherence to the ABC pathway in a general AF population using ideal definitions of ABC adherence. This variation in definitions and criteria included also precluded any attempts to combine the results of individual studies in a meta-analysis.

Conclusion
All studies consistently showed statistically significant reductions in the risk of stroke, myocardial infarction, and mortality among those with treatment adherent to the ABC pathway. The ABC pathway provides a simple decision-making framework to enable consistent equitable care from clinicians in both primary and secondary/tertiary care. Further research examining the impact of ABC pathway implementation in prospective cohorts where consistent inclusion criteria and definitions of 'A', 'B', and 'C' adherent care can be used is needed.

Supplementary material
Supplementary material is available at Europace online.