Factors associated with warfarin discontinuation, including bleeding patterns, in atrial fibrillation patients

Abstract

Objective:

To determine risk factors for both warfarin discontinuation and bleeding in patients with atrial fibrillation (AF).

Methods:

Data from the MarketScan database (January 2005–June 2008) were retrospectively analyzed for patients ≥18 years old who used warfarin continuously (≥2 prescriptions for 6 months) subsequent to an AF diagnosis. Patients were followed until one of the following endpoints occurred: warfarin discontinuation or end of the 30 month study period (whichever happened first). Recent bleeding was defined as occurring within 90 days before discontinuation. Drug interactions related to bleeding were defined as occurring within 120 days prior to bleeding.

Results:

The study included 7971 eligible patients (mean age 67.8 years; 41.2% female). During follow-up, 51.7% of patients discontinued warfarin (P < .001). More patients with recent bleeding (61.8%) discontinued warfarin compared to patients without recent bleeding (51.3%). After adjustment, patients with recent bleeding were 35% more likely to discontinue warfarin compared to those without recent bleeding (relative risk = 1.35; CI: 1.16–1.58). Age >85 years and a history of ≥1 hospitalizations/emergency room visits were associated with an increased likelihood of discontinuation (P < .001). Female gender, daily dosage >5 mg, concomitant use of gastroprotective agents, and CHADS₂ scores ≥1 were associated with a decreased likelihood of discontinuation (P < .05).

Conclusions:

Risk factors for warfarin discontinuation include older age, recent bleeding, and a high number of concomitant medications. Risk factors associated with bleeding events are older age, use of a warfarin-potentiating medication, previous bleeding, and higher CHADS₂ scores.

Keywords::

• Anticoagulants
• Atrial fibrillation
• Atrial flutter
• Bleeding
• Warfarin discontinuation

Transparency

Declaration of funding

This study was partially funded by a research grant from Chung-Ang University and an unrestricted research grant from Janssen Scientific Affairs, LLC.

Declaration of financial/other relationships

DS and SK have no conflicts of interest to declare. JCC, WN and JS have ownership interests (stock) in Johnson & Johnson (J&J). JCC is an employee of LifeScan, Inc. (a J&J company). WN and JS are employees of Janssen Scientific Affairs, LLC. CMRO Peer Reviewers on this manuscript have no relevant financial relationships to disclose.

CMRO peer reviewers may have received honoraria for their review work. The peer reviewers on this manuscript have disclosed that they have no relevant financial relationships.

Acknowledgments

The authors thank In-sun Choi PhD for data analysis and assistance with manuscript preparation. Editorial support was provided with funding from Janssen Scientific Affairs, LLC.