Contemporary clinical and economic outcomes among oral anticoagulant treated and untreated elderly patients with atrial fibrillation: Insights from the United States Medicare database.

Background: Oral anticoagulants (OACs) mitigate the risk of stroke in atrial fibrillation (AF) patients.
Objective: Elderly AF patients who were treated with OACs (apixaban, dabigatran, edoxaban, rivaroxaban, or warfarin) were compared against AF patients who were not treated with OACs with respect to their clinical and economic outcomes.
Methods: Newly diagnosed AF patients were identified between January 2013 and December 2017 in the Medicare database. Evidence of an OAC treatment claim on or after the first AF diagnosis was used to classify patients into treatment-defined cohorts, and these cohorts were further stratified based on the initial OAC prescribed. The risks of stroke/systemic embolism (SE), major bleeding (MB), and death were analyzed using inverse probability treatment weighted time-dependent Cox regression models, and costs were compared with marginal structural models.
Results: The two treatment groups were composed of 1,421,187 AF patients: OAC treated (N = 583,350, 41.0% [36.4% apixaban, 4.9% dabigatran, 0.1% edoxaban, 26.7% rivaroxaban, and 31.9% warfarin patients]) and untreated (N = 837,837, 59.0%). OAC-treated patients had a lower adjusted risk of stroke/SE compared to untreated patients (hazard ratio [HR]: 0.70; 95% confidence interval [CI]: 0.68-0.72). Additionally patients receiving OACs had a lower adjusted risk of death (HR: 0.56; 95% CI: 0.55-0.56) and a higher risk of MB (HR: 1.57; 95% CI: 1.54-1.59) and this trend was consistent across each OAC sub-group. The OAC-treated cohort had lower adjusted total healthcare costs per patient per month ($4,381 vs $7,172; p < .0001).
Conclusion: For the OAC-treated cohort in this elderly US population, stroke/SE and all-cause death were lower, while risk of MB was higher. Among OAC treated patients, total healthcare costs were lower than those of the untreated cohort.
Competing Interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: MBM and HY have no financial relationships or other potential conflicts of interest to declare. AK and RM are paid employees of STATinMED Research, which is a paid consultant to Bristol Myers Squibb and Pfizer. in connection with the development of this manuscript. PH, CR, BE, and MC are paid employees of Pfizer. JDG and MF are paid employees of Bristol Myers Squibb. JCH has received research support from Janssen Pharmaceuticals, Pfizer, and Bristol Myers Squibb.