In their paper on the safety of antithrombotic therapy for atrial fibrillation, Burton and colleagues point out that patients with atrial fibrillation in the community are older and have more comorbidity than those included in trials.1 This is very topical as UK GPs go through their registers of patients with atrial fibrillation with a view to maximising their Quality and Outcomes Framework points and related income in 2007.
In August 2006 we examined the medical records of all 43 patients known to have persistent atrial fibrillation in our practice. As in Burton's study, the mean age of the patients was 77 years (range = 56–94 years), but five (12%) were of African origin reflecting our inner city population. Eighteen patients (42%) were on warfarin, 16 on aspirin, two on clopidogrel and seven did not seem to be taking any antithrombotic treatment of whom four were in a nursing home. A questionnaire survey found that only 13 (59%) of 22 responders knew that warfarin or aspirin helps to prevent strokes or blood clots.
As in Burton's study we noted a ‘healthy user effect’ in warfarin prescribing. Of patients aged <75 years, 71% (12/17) were prescribed warfarin compared with 23% (6/26) of those aged 75 years (P<0.01). However, despite the tendency for GPs to prescribe warfarin for their fitter, younger patients, Burton and colleagues found that the risk of severe bleeding on warfarin, defined as death, intracranial bleeding or hospital admission, was 2.6% per patient year, twice the rate found in clinical trials. Over a fifth of patients on warfarin consulted their GP with at least one bleeding event during up to 5 years follow up. Combining these risks with the additional effort involved in anticoagulant monitoring, it is not surprising if compared to hospital doctors, many GPs have a more conservative approach to starting patients with atrial fibrillation on warfarin.
Acknowledgments
We are very grateful to the patients and staff at the Curran Practice, Manor Health Centre, London SW4 ORE.
Notes
Competing interests
The authors have stated that there are none.
- © British Journal of General Practice, 2006.