Atrial fibrillation (AF) is a common arrhythmia in older age and is an important risk factor for stroke. Anti-thrombotic therapy can prevent strokes and TIA in patients with chronic atrial fibrillation. A recent community-based study has shown that General Medical Services (GMS) prescriptions for digoxin is a reliable surrogate for AF [1], with a positive predictive value of 74% and a high specificity of 99.8% [2].

The aim of our study was to determine if the use of warfarin and aspirin therapy in GMS patients in Ireland with atrial fibrillation could be developed as an indicator of quality in community prescribing.

From the National General Medical Services Prescription Database for Ireland (≈1.3 million, approximately 30% of the population) for the year 2003, we identified patients over the age of 45 years who had received three or more prescriptions of digoxin. Medicines were coded using the WHO Anatomical Therapeutic Chemical (ATC) classification system. The use of warfarin, aspirin and the co-prescription of warfarin and aspirin in any 1 month was examined for the whole population. We also divided patients receiving digoxin in the three age bands based on the current guidelines [3] for anti-coagulation in chronic atrial fibrillation i.e., under 65 years, between 65-75 years and over 75 years. Data were analyzed using SAS version 8. Differences between groups were analyzed using chi square test. Logistic regression was used to calculate odds ratio (OR) and 95% confidence intervals. P < 0.05 was considered significant.

We identified 27 971 patients receiving prescriptions for digoxin, 51% females. Overall, 36% were on warfarin, 51% on aspirin and 8% on both warfarin and aspirin in any 1 month, while 5% were on no such treatment. Proportionately, significantly more men were on warfarin compared with women (40% vs 33%, p < 0.0001), though aspirin prescribing was quite similar (50 % vs 52%, p < 0.05). In the over 75 year age group (n = 18,966), only 29% of patients were on warfarin and 54% on aspirin. Again, fewer women than men received warfarin (27% vs 33%, p < 0.0001) while prescribing for aspirin was similar for men and women respectively (4% vs 53%). People over 75 years were three times less likely to receive warfarin than the under 65 year olds (OR 0.32, CI 0.30, 0.36, p < 0.0001). The odds ratio for receiving warfarin for women compared with men was 0.84 (CI 0.79, 0.88, p < 0.0001). The over 75 year olds were 1.5 times more likely to receive aspirin than the under 65 year olds (OR 1.76, CI 1.6, 1.93, p < 0.0001) and again women were prescribed less aspirin than men (OR 0.89, CI 0.84, 0.93, p < 0.0001).

There is evidence of low rates of prescribing for warfarin and aspirin in patients with AF, particularly in the elderly, where the risk of stroke secondary to AF is the highest. Women receive disproportionately less preventive therapy. These data suggest a considerable missed opportunity to prevent stroke. The demonstration of age and gender inequality suggests that the study of anti-thrombotic therapy in patients receiving digoxin may be used as a prescribing quality indicator.

1. White S, et al. IR Med J 2004; 97: 10.
2. O’Dubhghaill CT, et al. IR Med J 2003; 96: 296.
3. Stanley G, et al. JACC 2004; 43: 929.