Does the Birmingham model of oral anticoagulation management in primary care work outside trial conditions?

The effectiveness of the Birmingham model of primary care oral anticoagulation management has previously been demonstrated within a randomised controlled trial. The aim of this study was to assess the effectiveness of the Birmingham model in routine care. All patients from 12 primary care centres attending either practice-based or hospital-based anticoagulation clinics were retrospectively followed up from October 1996 to March 1998. Outcome measures were therapeutic International Normalised Ratio (INR) control, haemorrhagic and thrombotic episodes, and recall frequency; 452 patients who had two or more INR results during the follow-up period were investigated. There were no significant differences between practice-based and hospital-based populations in terms of the percentage time in range, (69% and 64% respectively). The proportion of tests in range was significantly higher in the practice-based group (61% practice-based, 57% hospital-based; P = 0.015). There was no difference between the two populations in terms of mean follow-up time (36 days in each group). There were no significant differences between groups for the number of clinical outcomes per patient. This study confirmed that, within these practices, oral anticoagulation management is safe and effective using the Birmingham model.