Background: Self-management may be an option to monitor oral anticoagulant therapy in health systems, but before recommending it, we need to assess patients’ ability to take on this task. The purpose of the study was to describe patients’ ability to self-manage and associated factors. Methods: This was a 3-year prospective quasi-experimental study with a control group. Overall, 333 patients on anticoagulant therapy from seven primary care health centres of the Basque Health Service were included in the intervention group and followed up for 6 months after the intervention, assessing their ability to self-test and self-manage. The intervention consisted of a patient training programme, providing detailed information on their condition and its treatment, and practical training in how to use a portable blood coagulation monitor and adjust their anticoagulant dose. Comparisons were made with a control group (333 patients receiving OAT under usual care from the same seven health centres). Outcome variables were ability to self-manage, quality of the outcome (in terms of time in therapeutic range), and quality of life in the intervention group, and general patient characteristics (age and sex), clinical variables (reason for OAT, INR range), and quality of the outcome (in terms of percentage of INR measurements in range and complications) in both groups. Results: Overall, 26.13 % of patients invited to participate in the intervention agreed. Of these, 99 % successfully learned to self-manage their OAT. Just 4.2 % did not complete the follow-up, in all cases for reasons unrelated to self-management, and 4.5 % required additional learning support. Outcomes were better than under usual care in terms of percentage of INR measurements in range (12 %), rate of complications (4 %) and quality of life (9.2 %). Limitations: Patients were only followed-up period for 6 months and the study was conducted in a single health organization. Though patients eligible to participate were selected randomly, they were not randomly allocated to the groups. This is a potential source of selection bias. Data needed to calculate in-range time were not collected from controls; rather the results for the self-management group were compared with external data from other studies. Conclusions: Almost all participants achieved competency in self-management, with no differences by age, sex, concurrent illnesses, polypharmacy or educational level. The greatest barrier to self-management was the attitude of patients themselves and those around them. Self-management in primary care is a good alternative to usual care, patients having longer times in therapeutic range and fewer complications, and improving their quality of life. Remote management is a good support tool.