Using a personalized decision support algorithm for dosing in warfarin treatment: A randomised controlled trial

Vitamin K-antagonists such as warfarin treatment remain the mainstay to prevent thromboembolic events in various conditions. The quality of the treatment is reflected through time in therapeutic (INR) range (TTR) with a threshold at ≥ 70% indicating ‘good quality’; achieving this quality is not trivial. We conducted a randomised controlled trial to assess the impact of decision aiding model on treatment quality in a high quality vitamin K-antagonist treatment setting. We investigated if algorithm-suggested warfarin dosing was superior to standard dosing in a high-quality setting involving self-managing warfarin patients. Patients were initially allocated to either algorithm-suggested warfarin dosing or to standard care treatment, and were crossed over after three months. The trial period was a total of six months, and the primary endpoint was TTR; we also investigated a secondary endpoint of log-transformed INR variability. A total of 191 patients contributed to the main analysis with a mean follow-up time of 140 days; 75% were males and the mean age was 65 years old. The intervention arm achieved a TTR of 81.6, while the placebo arm attained a TTR of 80.9 (difference intervention arm minus placebo arm: 0.67 (95% confidence interval − 2.93 to 4.27). The difference in INR variability was 0.30 (0.14 to 0.47), favouring the placebo arm in terms of lower log transformed variability. We found no difference between the two trial-arms in a high-quality warfarin treatment setup. However in general, the model performed similarly as to routine patient self-management care. (ClinicalTrials.gov number: NCT02705976)