• atrial fibrillation

Atrial fibrillation (AF) is a condition of global importance with significant and progressive effects on morbidity, mortality and healthcare expenditure.1 The prevalence of AF is rising, and it is projected to at least double over the next 30 years due to advancing age and increasing risk factors for developing AF including cardiovascular illness and adverse lifestyle factors. AF is the leading cause of stroke, and by comparison to other types of stroke,2 AF-related stroke is more severe; yet highly preventable. Although we have overwhelming evidence that oral anticoagulation (OAC) is highly effective in preventing stroke in patients with AF,3 therapy has remained underused until recently.

In their Heart paper, Adderley et al4 report the results of temporal trends in age-sex standardised AF prevalence and use of stroke prevention therapy among 744 primary care practices across the UK from 2000 to 2016. Once again, this demonstrates the value of high-quality (and very large) data sets utilising data from a practice rather than a hospital level, the authors are able to characterise over 5 million individuals with AF. Seventeen sequential cross-sectional analyses were performed using The Health Improvement Network database, to identify 5 058 699 patients with AF, characterising stroke risk profile and use of antithrombotic therapies. A particular strength of this study is the primary care setting as the increasing burden of AF means primary care physicians are likely to manage more AF.5 Older studies6 7 predominately from the UK provide some of the only estimates on the contribution of primary care cases attributable to AF, which had been increasing, and this current study provides much needed contemporary data. Additional strengths include their ability to capture antiplatelet agents including aspirin and identify contraindication to antithrombotic therapy as determined by the patient’s primary care physician, both are commonly absent from administrative database studies.

Among eligible patients with AF in primary care practices, rates of OAC increased from 35% to 76% for those at high risk for stroke (CHA₂DS₂VASC>2) and from 33% to 47% for those at moderate risk of stroke (CHA₂DS₂VASC=1 and male) while overtreatment with OAC decreased to 10% in low-risk patients. Across all stroke risk strata, use of antiplatelet agents decreased.

How should these data be viewed? This should be seen as a remarkable transformation in stroke prevention across all stroke risks compared with prior studies in the UK and elsewhere.8 9 However, further work is needed to transition patients to direct OAC (DOAC) therapy as only 38% of patients received a DOAC in 2016 despite international guidelines and an evidence base exceeding 80 000 patients showing the more effective stroke prevention with an improved safety profile over that of warfarin. The availability of DOAC therapy likely had a major impact on patient and physician decision to initiate stroke prevention therapy. The authors do not report on the trends in OAC prescription and the type of agent for patients with AF >85 years, an age group which represented the largest portion of the increasing AF prevalence and where DOAC may have a distinct advantage over warfarin. Clinicians may overestimate perceived fall or bleeding risk in the elderly and whether notion persists in the setting of DOAC therapies is unclear.

Another important explanation for the increased rates of appropriate OAC use is the implementation of a scheme to financially incentivise AF management. Although an effective strategy, this approach requires a comprehensive economic evaluation, lobbying from healthcare professionals and buy-in from policymakers which may not be feasible in other countries. The authors also mention changes to European guidelines as a possible reason for improved use of OAC and substantial drop in antiplatelet prescribing; however, these publications alone minimally influence clinical practice.10 Understanding clinician factors (eg, number of years in practice, number of clinicians in the practice and other quality improvement efforts) within the primary care practices will aid in the efforts for knowledge translation. Integration of guideline recommendations within clinical flow, that is, electronic flags in medical records or standardised order forms, may allow for more evidence-based practice perhaps focused on the lowest preforming quintile. Finally, the unintended consequences of focusing on AF must be understood—did hypertension control, lipid or diabetes management lose momentum? Panvascular risk reduction will aid the UK realise its overall cardiovascular and stroke reduction goals.

In this study, Adderley et al4 address an important gap regarding use of stroke prevention therapy for patients with AF in primary care, a setting which will likely continue to encounter a growing AF burden. Although it is encouraging, appropriate use of antithrombotic therapy is increasing, additional simple and cost-effective strategies are needed to demonstrate an even higher use of DOAC in eligible patients, particularly those at moderate risk, minimise overtreatment and reduce practice variations that can be implemented in different healthcare systems where incentivising may not be an option.