Abstract 1575: Atrial Fibrillation and Warfarin: Time in Therapeutic Range – an Electronic Medical Record System Study of Real-World Practice
Purpose: A study of longitudinal INR levels among a cohort of patients with atrial fibrillation and/or flutter (AF) could enhance understanding of time-in-therapeutic-range and the effectiveness and risks of warfarin.
Methods: We searched the Regenstrief Medical Record System for patients > 18 years old with new onset AF from 1998 to 2007. The first AF diagnosis was defined as the “AF index date.” Patients with mitral or aortic valve replacement, hyperthyroidism, or no clinical visit within one year after the AF index date were excluded. Warfarin exposure was defined by electronic pharmacy records or physician order entry. The patients’ data were followed longitudinally (whether or not continuously on warfarin) from the first warfarin exposure after the AF index date until either the last INR or 30 days after the last warfarin exposure date, whichever was later. Patients’ first stroke or first gastro-intestinal bleeding event was identified by ICD9 codes. INR level was assumed to increase or decrease linearly over time between measurements (Rosendaal method). If no INR was recorded on or before the first warfarin date, a value of 1.0 was assigned for that date, with which to connect linearly with the next INR. Similarly, thirty days after the last warfarin date (or the last INR, whichever was later), the INR value was set equal to 1.0.
Results: Among 3329 subjects who met criteria for AF, 1502 (mean age 64 years, s.d. 13; 56% men) had at least one prescription or order for warfarin on or after the AF index date. Seventeen patients with no INR values were excluded. Time in INR ranges is shown in the Table⇓. First stroke and GI bleed events, and incidence rates per patient year are presented.
Conclusion: In this study of real-world practice, patients with AF were often not in therapeutic range; stroke and GI bleeding rates were lowest in that range. These results help depict the relationship between two clinical outcomes and warfarin/INR “in range” exposure.