Of 125 patients aged 65 years or over, with atrial fibrillation taking warfarin for at least 12 months, with a standard deviation (SD) of prothrombin time, expressed as the International Normalized Ratio (INR) >0.5 over the previous 6 months, 40 were randomized to continue with usual clinic care and 85 to receive education about warfarin. Of these, 44 were randomized to self-monitor their INR and 41 returned to clinic. Compared with the previous 6 months there was a significant increase in percentage time within the therapeutic range for the 6 months following education [61.1 vs. 70.4; mean difference 8.8; 95% confidence interval (CI): -0.2-17.8; P = 0.054] and following education and self-monitoring (57 vs. 71.1; mean difference 14.1; 95% CI: 6.7-21.5; P < 0.001), compared with those patients following usual clinic care (60.0 vs. 63.2; mean difference 3.2; 95% CI: -7.3-13.7). Using the same comparative periods, the INR SD fell by 0.24 (P < 0.0001) in the group allocated to education and self-monitoring, 0.26 (P < 0.0001) in the group receiving education alone and 0.16 (P = 0.003) in the control group. Inter-group differences were not statistically significant (intervention groups 0.26 +/- 0.30 vs. control 0.16 +/- 0.3, P = 0.10). Quality-of-life measurements and health beliefs about warfarin were unchanged (apart from emotional role limitation) with education or education and self-monitoring. Patient education regarding anticoagulation therapy could be a cost-effective initiative and is worthy of further study.