Background: Practice guidelines recommend long-term stroke prophylaxis in patients with chronic atrial fibrillation (cAF).
Objectives: To examine treatment initiation and persistence and factors that influence the choice of cAF treatment.
Patients/methods: This study used the General Practice Research Database, including computerized medical records of general practitioners in the UK. Patients aged 40+ years with cAF after 1 January 2000 were included. Cox proportional hazards regression models evaluated initiation and treatment continuation over time of warfarin and aspirin. Treatment discontinuation was defined as no repeat prescription within a three-month period after the expected end of the treatment course.
Results: The study population included 41 910 cAF patients. Elderly patients (aged 85+) were less likely to start warfarin [relative rate (RR) = 0.16, 95% confidence interval (CI) 0.15-0.18] and more likely to start aspirin (RR = 1.66, 95% CI 1.47-1.88) than patients aged 40-64 years. A history of dementia (RR = 0.28, 95% CI 0.17-0.44) and falls (RR = 0.76, 95% CI 0.70-0.83) also reduced the likelihood of warfarin initiation. Adjusting for age and gender, higher stroke risk (CHADS2 score) was not found to be associated with initiation of warfarin or aspirin contrary to current guidelines recommendations. One-year persistence was 70% for warfarin and 50% for aspirin. Treatment persistence was higher in elderly patients using warfarin and aspirin. A higher CHADS(2) score was associated with improved persistence only with warfarin.
Conclusions: The low likelihood of patients with cAF in general practice remaining on treatment long-term indicates that not all benefits as observed in clinical trials may be achieved in usual clinical practice.