Time Trends and Treatment Pathways in Prescribing Individual Oral Anticoagulants in Patients with Nonvalvular Atrial Fibrillation: An Observational Study of More than Three Million Patients from Europe and the United States

Abstract

Background: Data directly comparing trends in the use of different oral anticoagulants (OACs) among patients with atrial fibrillation (AF) from different countries are limited. We addressed this using a large-scale network cohort study in the United States (US), Belgium, France, Germany, and the United Kingdom (UK).

Methods: We used nine databases (claims or electronic health records) that had been converted into the Observational Medical Outcomes Partnership Common Data Model with analysis performed using open-source analytical tools. We identified adults with AF and a first OAC prescription, either vitamin K antagonist (VKA) or direct oral anticoagulant (DOAC), from 2010 to 2017. We described time trends in use, continuation, and switching.

Results: In 2010, 87.5%-99.8% of patients started on a VKA. By 2017, the majority started on a DOAC: 87.0% (US), 88.3% (Belgium), 93.1% (France), 88.4% (Germany), and 86.1%-86.7% (UK). In the UK, DOACs became the most common starting OAC in 2015, 2-3 years later than elsewhere. Apixaban was the most common starting OAC by 2017, 50.2%-57.8% (US), 31.4% (Belgium), 45.9% (France), 39.5% (Germany), and 49.8%-50.5% (UK), followed by rivaroxaban, 24.8%-32.5% (US), 25.7% (Belgium), 38.4% (France), 24.9% (Germany), and 30.2%-31.2% (UK). Long-term treatment was less common in the US than in Europe, especially the UK. A minority of patients switched from their index OAC in the short and long term.

Conclusions: From 2010 to 2017, VKA use had significantly declined and DOAC use had significantly increased in the US and Europe. Apixaban was the most prescribed OAC in 2017, followed by rivaroxaban.

Figure 1

The frequency distribution of each index OAC for each database across the study period (patients with AF).

Figure 2

Frequency distribution of each index OAC for each calendar year and first OAC prescribed in 2017 (patients with AF still available for observation in 2017).

Figure 3

Two-year OAC treatment sequences (irrespective of gaps in treatment) in patients with AF by database and study year (patients still available for observation at 2 years after their index prescription). Note: the inner circle of each sunburst plot shows the percentage of patients prescribed each OAC type (first OAC prescription) in that year, coloured segments in the next outer circle show the second OAC prescribed (if any) at any time during the two-year follow-up period (i.e., the first OAC switch), and coloured segments in the second outer circle (if any) show the third OAC prescribed (if any) at any time during the two-year follow-up period (either a switch back to the original OAC prescribed or a switch to another different OAC. Also, very few patients were available for analysis in the US Longitudinal Prescription Diagnosis database and the French Disease Analyzer database in 2010. Hence, results are not shown for these databases in this year. Each cell in the sunburst plot represents 2%.