ABSTRACT: Importance:Less than half of US patients with a diagnosis of atrial fibrillation (AF) receive oral anticoagulation. Objectives:To identify patients with similar patterns of adherence to regimens of warfarin and direct oral anticoagulants (DOACs) in the first year after AF diagnosis and to evaluate associations between patient characteristics and membership in latent classes of adherence. Design, Setting, and Participants:This retrospective cohort study used 2013 to 2016 Medicare claims data to identify 7491 patients with a new diagnosis of AF in 2014 to 2015 who initiated warfarin after AF diagnosis and 9478 patients with a new diagnosis of AF in 2014 to 2015 who initiated DOAC treatment after AF diagnosis, for a total of 16?969 Medicare beneficiaries. Participants were followed up for 12 months after AF diagnosis. Statistical analysis was performed from February 1 to November 30, 2018. Exposures:Treatment with warfarin or DOAC after AF diagnosis. Main Outcomes and Measures:The main outcome was the proportion of days that patients received warfarin or DOAC, measured in 30-day intervals after AF diagnosis. Independent variables included patient demographic characteristics, socioeconomic status, region of residence, and clinical characteristics. Latent class mixed models were used to identify latent classes of warfarin and DOAC adherence, and polytomous logistic regression was used to assess the association between patient characteristics and membership in each latent class. Results:Among the 7491 patients receiving warfarin (4348 women), the mean (SD) age was 76.0 (10.0) years; among the 9478 patients receiving DOAC (5496 women), the mean (SD) age was 77.0 (8.5) years. Four latent classes of patients were identified based on warfarin adherence: late initiators (980 [13%]), early initiators who discontinued therapy at months 1 to 3 (1297 [17%]) or at months 5 to 10 (735 [10%]), and continuously adherent patients (4479 [60%]). Four latent classes of patients were also identified based on DOAC adherence: patients who initiated DOAC in months 1 to 5 (1368 [14%]) or months 6 to 11 (800 [8%]), patients with suboptimal and decreasing adherence (2267 [24%]), and continuously adherent patients (5043 [53%]). Membership in latent classes of warfarin adherence was significantly associated with sex, eligibility for Medicaid and income subsidy, region of residence, CHA2DS2-VASc (cardiac failure or dysfunction, hypertension, age 65-74 [1 point] or ?75 years [2 points], diabetes, and stroke, transient ischemic attack or thromboembolism [2 points]-vascular disease, and sex category [female]) risk score, and HAS-BLED (hypertension, abnormal renal and liver function, stroke, bleeding, labile international normalized ratio, elderly, and drugs or alcohol) score. Membership in latent classes of DOAC adherence was significantly associated with race/ethnicity, region of residence, HAS-BLED score, and use of antiarrhythmic medications. Conclusions and Relevance:This study found that, among patients who initiated anticoagulation therapy, 40% of those who initiated warfarin therapy and 47% of those who initiated DOAC treatment did not continuously adhere to therapy in the first year after AF diagnosis. Identifying longitudinal patterns of warfarin and DOAC adherence and the factors associated with them provides suggestions for the design of targeted strategies to mitigate suboptimal oral anticoagulation use.
