Adverse Outcomes and Predictors of Underuse of Antithrombotic Therapy in Medicare Beneficiaries With Chronic Atrial Fibrillation

Abstract

Background and Purpose —Antithrombotic therapy can prevent strokes and transient ischemic attacks (TIAs) in carefully selected patients who have chronic nonvalvular atrial fibrillation (NVAF). Our objectives were 3-fold: to document the use of warfarin and aspirin therapy in Missouri Medicare beneficiaries with chronic NVAF; to identify factors associated with warfarin and aspirin underuse; and to determine the association between prescription of warfarin and aspirin at hospital discharge and adverse outcomes in this elderly, frail population. Methods —We linked chart reviews from all Missouri hospitals to Medicare claims data from 1993 to 1996. From chart reviews, we documented Medicare beneficiaries’ demographic factors, comorbid conditions, and antithrombotic therapy prescribed at the time of hospital discharge. From Medicare claims, we determined the date of outcomes—death from any cause or hospitalization for an ischemic event (a stroke, a TIA, or a myocardial infarction). Results —Only 328 (55%) of the 597 Medicare beneficiaries were prescribed antithrombotic therapy at hospital discharge: 34% received warfarin and 21% received aspirin. Advanced age, female gender, and rural residency predicted underuse of antithrombotic therapy. After controlling for these factors, as well as stroke risk factors and contraindications to anticoagulation, the prescription of warfarin was associated with a 24% relative risk reduction (RRR) in adverse outcomes ( P =0.003). Prescription of aspirin was associated with a nonsignificant 5% RRR in these events ( P =0.56). Conclusions —The underuse of antithrombotic therapy in Medicare beneficiaries who have NVAF is associated with measurable adverse outcomes. The benefit of warfarin therapy may extend to frail, elderly patients, a group that was excluded from randomized controlled trials. The role of antiplatelet therapy in this population deserves further study because many of these patients have relative contraindications to warfarin.