Background and Objectives: Anticoagulation clinics are widely used for anticoagulation management in many countries, but have only recently began to gain acceptance in Taiwan. Our service model is a physician-managed outpatient clinic collaborating with clinical pharmacist and nurse. This study aimed to evaluate the adequacy of anticoagulation and rates of warfarin-related complications before and after referral to our collaborative anticoagulation clinic (CAC). Methods: Stroke patients taking warfarin from the neurology department were identifi ed and referred to the CAC during the 12-month period from February 2009 to January 2010. Quality markers include percentage of international normalized ratio (INR) values in the therapeutic range, frequency of INR monitoring, and frequency of follow-up visits and the mean interval of next INR monitoring after non-therapeutic INRs were compared one year before and after management in the CAC. Using studied patients as self-control, they were included in the analysis if patients had at least 3 months follow-up or 3 INR values both before and after referral. Results: A total of 44 stroke patients were included: mean age of 75.0 ± 9.7 years, with a CHADS2 score of 3.71 ± 0.69. The adequacy of anticoagulation was signifi cantly greater during CAC care compared with the period before referral; the percentage of INR within expanded therapeutic range was 60.9% versus 53.7%, respectively (p=0.049). Reduction in sub-therapeutic INR values from 31.8% to 24.2% (p=0.023) contributed mostly to the improved quality of care. The time interval of next INR monitoring after non-therapeutic INRs ( 4.0 or 1.5) was also signifi cantly shorter. However, there was no signifi cant difference in the rates of thromboembolic and hemorrhagic events which may be attributed to a small sample size. Conclusion: Based on results of our study, a CAC may be the optimal structure for anticoagulation management service in the future.