Effectiveness and safety of self-managed oral anticoagulant therapy compared with direct oral anticoagulants in patients with atrial fibrillation: a propensity-weighted cohort study

Erik Lerkevang Grove, Flemming Skjoeth, Peter Brønnum Nielsen, Thomas Decker Christensen, Torben Bjerregaard Larsen

Research output: Contribution to journalConference abstract in journalResearchpeer-review

Abstract

We compared the effectiveness and safety of direct oral anticoagulants (DOAC) vs patient self-managed warfarin therapy (PSM) in patients with atrial fibrillation. We linked prospectively registered data from university hospital clinics to nationwide Danish health registries. Primary effectiveness and safety outcomes were ischaemic stroke (incl. systemic embolism) and major bleeding. All-cause mortality and all-cause stroke were secondary outcomes. An inverse probability of treatment propensity-weighted approach was applied to adjust for potential confounding. The study cohorts included 534 patients treated with PSM and 2,671 patients treated with DOAC. Weighted rates of ischaemic stroke were 0.46 and 1.30 percent per year with PSM vs DOAC, hazard ratio (HR) 0.27 (95% confidence interval 0.11-0.68) with 2.5 years follow-up. Rates of major bleeding were 2.32 and 2.13 percent per year (HR 1.06 [0.69-1.63]). All-cause mortality was not statistically different (HR 0.67 [0.39-1.17]), whereas the incidence of all-cause stroke was significantly lower among patients treated with PSM with rates of 0.61 vs 1.45 percent per year (HR 0.36 [0.16-0.78]). In patients with atrial fibrillation, self-managed oral anticoagulant treatment was associated with a significantly lower risk of all-cause and ischaemic stroke compared to treatment with DOAC, whereas no significant differences were observed for major bleeding and mortality.

Fingerprint

Dive into the research topics of 'Effectiveness and safety of self-managed oral anticoagulant therapy compared with direct oral anticoagulants in patients with atrial fibrillation: a propensity-weighted cohort study'. Together they form a unique fingerprint.

Cite this