BACKGROUND: Patient self-management (PSM) of anticoagulant treatment with vitamin K antagonist (VKA) has emerged as an effective approach for maintaining the international normalized ratio (INR) within the therapeutic range. The objective of this quality assurance project, conducted in clinical practice, was to evaluate the long-term effectiveness and safety of anticoagulant treatment with warfarin during PSM compared to conventional treatment administered by general practitioners (GPs). METHODS: This cohort study, using a retrospective and prospective design, included 400 patients who underwent PSM training for a 21-week period between 2011 and 2020. Clinical data extracted from the patient journal systems included hospitalization due to severe clinical complications. The primary outcome was any difference in the yearly risk of hospitalization between the conventional and PSM periods. Secondary outcomes included variations in time within the therapeutic range (TTR), INR fluctuations, and incidence of extreme INR values. RESULTS: The median treatment duration was 2.45 years (25th-75th percentile 0.80, 7.35) for the conventional period and 4.99 years (25th-75th percentile 2.41, 7.43) for the PSM period. The annual risk for hospitalization due to severe bleeding was 1.25% during PSM compared to 1.69% during conventional treatment (p = 0.885). The yearly risk for hospitalization due to thrombosis was 0.67% during PSM versus 1.48% during conventional treatment (p = 0.256), and the annual risk for hospitalization due to spontaneous bleeding, thrombosis, or thromboemboli was 1.12% versus 2.76% (p = 0.112). Median TTR (25th-75th percentile) increased from 71.6% (60.0, 82.7) to 78.6% (67.9, 91.7) (p <
0.001), while INR variance decreased from 21.0% to 16.5% (p <
0.001). The proportion of extreme subtherapeutic INR values (≤ 2.0 (≤ 1.5 for patients with mechanical ON-X aortic valve prostheses)) decreased from 14.0% to 5.0% (p <
0.001) during PSM, whereas the proportion of high-level INR (≥ 5.0) remained unchanged (0.6%). CONCLUSIONS: The long-term evaluation of PSM of warfarin treatment in clinical practice suggests that PSM for suitable patients selected by GPs is as safe as conventional GP treatment.