All data were deidentified and the study was exempt from review by the Institutional Review Board of the Centers for Disease Control and Prevention.

Data were extracted from the Merative MarketScan Commercial Claims and Encounters, Medicare Supplement, and Medicaid databases from January 1, 2018 to June 30, 2021.⁵ Patients with a new diagnosis of AF from January 1, 2014, to December 31, 2017 (i.e., patient selection period), in the databases were selected. International Classification of Diseases, Ninth Revision, Clinical Modification of 427.3 (from January 2014 to September 30, 2015) and International Classification of Diseases, Tenth Revision, Clinical Modification of I48 (from October 1, 2015, to December 31, 2017) were used to identify AF diagnosis. Patients were identified as AF if there were at least one inpatient or emergency department visits or 2 outpatient visits at least 7 to 365 days apart from January 1, 2014, to December 31, 2017. We restricted patients to those continuously enrolled beginning 180 days before the date of first AF diagnosis to June 30, 2021. More details of the sample selection process and the corresponding diagnosis codes can be found in the supplementary materials and Appendix Table 1.

The differences in the proportions of patients with AF treated with each of the 5 FDA approved OACs (apixaban, dabigatran, edoxaban, rivaroxaban, or warfarin), and any OAC from January 2018 to June 2021 (December 2020 for patients in Medicaid) were tested using Welch’s 2-tail t test by insurance types. Summary statistics are provided of the medical costs and utilization of services and Charlson co-morbidity scores of AF patients with and without OAC prescription in 2019, the most recent year before the COVID-19 pandemic (the results are similar in other years and available upon request to the corresponding author). A p <0.05 indicates statistical significance. All analyses were conducted using Stata MP statistical software version 14.2 (StataCorp, College Station, Texas) in 2022 to 2023.

A total of 15,974 AF patients with commercial insurance (11,193 patients; 12% OAC users in January 2018), Medicare (1488 patients; 20% OAC users in January 2018), and Medicaid (3,293 patients; 15% OAC users in January 2018) were included and followed (Table 1). Compared to non-OAC users, OAC users had higher medical costs, healthcare utilization, and Charlson Comorbidity Index scores for all insurance types (Table 1). From January 2018 to June 2021, utilization of any OAC increased significantly (29.6% among commercially ensured, 48.8% for those in Medicare, and 31.5% for Medicaid enrollees). Apixaban use had the highest percentage increase (42.9% for commercial insurance, 66.5% for Medicare, 63.2% for Medicaid). At the end of follow-up period in 2021, the OAC utilization rates by insurance types ranged between 16% and 29% (Figure 1).

This study estimates long-term OAC utilization in patients with incident AF by 3 insurance types. OAC use increased in all insurance types, with the highest overall increase among Medicare beneficiaries. Among OACs, the highest increase occurred in apixaban use, which is consistent with previous trends during 2010 to 2017.³ Despite the significant increase in OAC use, utilization remained low. The drivers for the persistent underuse or delay in initiation require further research.

Our observed increases in apixaban use and decreases in warfarin and dabigatran use are consistent with several other previous studies in the United States.^3,6 These trends may be attributed to the American College of Cardiology and American Heart Association guideline recommendation of a direct OAC use over warfarin.¹

Limitations of this study include it is a descriptive study and the restriction to patients who were continuously covered by the same insurance during 2018 to 2021. This restriction also resulted in the exclusion of some groups of patients (e.g., uninsured and with capitated insurance plans).

Given the high burden of stroke in the United States and the strong evidence for OAC use in patients with AF and elevated stroke risk, patient, clinician, and health system interventions are needed to improve prescribing of and adherence to OACs.