Self-testing and clinical outcomes in Black and White patients on warfarin for atrial fibrillation or venous thromboembolism.

BACKGROUND

Warfarin management requires frequent international normalized ratio (INR) testing. Self-testing may reduce this burden and improve INR control and outcomes. However, concerns exist about variability in utilization and the effectiveness of self-testing across different patient groups.

OBJECTIVES

This study aimed to evaluate INR self-testing utilization and clinical outcomes in Black and White warfarin-treated patients.

METHODS

In this retrospective, observational study, Black and White patients who were using warfarin for atrial fibrillation or venous thromboembolism (April 2012 to July 2024) were identified from the Michigan Anticoagulation Quality Improvement Initiative registry. INR control and outcomes were compared between self-testers and non-self-testers. Rates were adjusted by inverse probability weighting; comparisons between racial groups used the negative binomial model. Major and nonmajor bleeding events were defined based on the International Society on Thrombosis and Haemostasis criteria. Moderation analysis examined whether race influences self-testing's safety and effectiveness.

RESULTS

Among 5903 warfarin-treated patients (20.5% Black, 79.5% White), self-testing was used by fewer than 1 in 7 patients (15.3% White, 10.1% Black). Self-testers had higher adjusted time in therapeutic range (TTR; 65.3% vs 59.8%; P < .001), fewer extreme INRs, and lower nonmajor bleeding rates (20.9 vs 29.9 per 100 patient-years; P < .0001) than patients who underwent traditional INR testing. Among Black patients, self-testers had higher TTR (58.8% vs 55.4%; P = .0018) and fewer nonmajor bleeds (23.1 vs 33.1 per 100 patient-years; P = .024) than Black patients who underwent traditional INR testing. Among White patients, self-testers had higher adjusted TTR (68.1% vs 61.3%, P < .0001) and fewer major (3.4 vs 4.9 per 100 patient-years, P = .014) and nonmajor bleeds (16.7 vs 26.2 per 100 patient-years, P < .0001) than White patients who underwent traditional INR testing. Thromboembolic events were similar between groups.

CONCLUSION

Self-testing was utilized infrequently, especially among Black patients, but was associated with better INR control and less bleeding than traditional INR testing. Increased utilization and support of self-testing may improve patient outcomes.