Abstract 9664: Utilization Patterns of Anticoagulants in Non-Valvular Atrial Fibrillation after the entry of Novel Oral Anticoagulants in the United States

Abstract

Introduction: Anticoagulants including warfarin and novel oral anticoagulants (NOACs) are recommended for stroke prevention in non-valvular atrial fibrillation (NVAF) patients. Limited data are available on real-world use of warfarin and NOACs in these patients since the entry of NOACs into the US market in October, 2010.

Hypothesis: Warfarin and NOACs are prescribed for stroke prevention differentially based on patient characteristics.

Methods: Incident cases of NVAF were identified from GE electronic medical records database between 11/2010 and 08/2011 based on ICD-9 codes 427.31 and 427.32. Baseline characteristics were obtained from medical record history prior to diagnosis. Patients were assessed over 6 months for initiation of warfarin or NOAC (dabigatran or rivaroxaban). Factors associated with anticoagulant initiation and NOAC initiation were examined using multivariate logistic regression analyses.

Results: Among 19,730 NVAF patients, mean age was 70.8 years, 45.5% were female, and 36.5% initiated an anticoagulant (31.0% warfarin and 5.5% NOAC). CHADS2 score differed across treatment groups with CHADS2 ≥2 found in 47.9%, 45.6% and 36.3% of patients receiving warfarin, NOACs, and no anticoagulants, respectively (p < 0.001). Age, prescriber type, hypertension, diabetes, CHF, prior stroke/TIA, prior MI, and history of GI or other bleeding were significantly associated with the use of anticoagulants. NOAC use was more likely than warfarin in patients with hypertension (OR=1.19), history of GI bleeding (OR=1.40) and prior aspirin use (OR=1.81) but less likely in those with history of CHF (OR=0.70).

Conclusions: Since the entry of NOACs, over 60% of newly diagnosed NVAF patients did not receive anticoagulants despite 36% of those patients having CHADS2 score ≥2. GI bleeding history, CHF, and prior aspirin use were major clinical factors associated with NOAC use. Further research is warranted to examine changes in anticoagulant treatment over time.