Abstract 18939: Trends in Antithrombotic Therapy for Atrial Fibrillation in the Veterans Affairs

Abstract

INTRODUCTION: The CHADS2 is a stroke risk score developed in 2001 to guide the selection of antithrombotic therapy for patients with atrial fibrillation (AF). Since 2001 the CHADS2 has been adopted into multiple practice guidelines. Anticoagulation by warfarin or newer anticoagulants is recommended for patients with CHADS2 of 2 or higher, with aspirin or clopidogrel recommended for lower CHADS2 scores. We evaluated trends during 2002-2011 in antithrombotic therapy by CHADS2 score for patients with new AF episodes in the Veterans Affairs (VA). METHODS: New AF episodes were defined by at least two encounters for AF (ICD-9-CM 427.31) and no previous AF episodes based on national VA data (n=297,611). Patient age and diagnoses for hypertension, diabetes, heart failure, and stroke during the 12 months preceding AF were used to calculate CHADS2 scores. Receipt of warfarin, dabigatran, or clopidogrel within 90 days of the AF episode was identified in VA national pharmacy data. Trends in antithrombotic therapy by CHADS2 score were examined in unadjusted analyses and in analyses that controlled for bleeding risk. RESULTS: Overall, 54% of patients received antithrombotics within 90 days of AF. Among those, 88% received warfarin and 18% received clopidogrel. Dabigatran (available in the VA in 2011) accounted for 0.4% of patients in 2010-2011. From 2002 through 2011, the proportion of patients prescribed warfarin within 90 days of AF decreased by 26%, 23%, 15%, 13%, 8%, and 10%, for patients with CHADS2 score of 0, 1, 2, 3, 4, and 5-6, respectively (p<.0010). Clopidogrel use decreased by 17% and 20% over the 10-year period for patients with CHADS2 of 1 or 5 or more, respectively, but increased for patients with CHADS2 of 1, 2, or 3. Results were similar controlling for bleeding risk factors. LIMITATIONS: We were unable to assess the use of aspirin due to its availability without a prescription. CONCLUSIONS: Widespread adoption of the CHADS2 stroke risk score in practice guidelines appears to have impacted anticoagulation prescribing patterns for warfarin. Results for clopidogrel were mixed, likely due to the use of clopidogrel for other cardiac conditions. The overall decline in antithrombotic agents in our study, even among patients with high CHADS, warrants investigation.