Background and purpose: Warfarin reduces the rate of stroke among patients with atrial fibrillation. We sought to determine warfarin use within a population sample of elderly patients with atrial fibrillation.
Methods: The Connecticut Peer Review Organization conducted a chart review of Medicare patients aged > or = 65 years with a history of atrial fibrillation before a hospitalization during the first 6 months of 1994.
Results: Among 488 patients (308 women; 457 white; 173 aged > or = 85 years), 38% (184/488) had a relative contraindication to anticoagulation (history of bleeding, dementia, alcohol use, falls, cancer, or the need for nonsteroidal anti-inflammatory drugs). Among the remaining patients (with known atrial fibrillation, but without a contraindication), only 38% (117/304) had been prescribed warfarin. Of those not prescribed warfarin, 63% (117/187) were also not taking aspirin. There were 272 patients with at least one additional vascular risk factor and no contraindication to anticoagulation. Among these patients at moderate to high risk for stroke, anticoagulation had been prescribed in 40% (109/272). Overal, among those not prescribed warfarin, 58% (95/163) were not taking aspirin. Patients admitted with a stroke were more likely to be significantly underanticoagulated (with international normalized ratio < 1.5) (43.5% versus 20.9% for those without stroke; P < .005). Anticoagulation was most effective for those with an international normalized ratio > or = 2.0.
Conclusions: Warfarin anticoagulation with atrial fibrillation, even among "ideal" candidates, appears dramatically underutilized. In addition, among those prescribed warfarin, patients are often undertreated. Increased warfarin use among patients with atrial fibrillation represents an excellent opportunity for stroke prevention in the elderly.