Quantifying the Cost of Warfarin-Related Bleeding Events in Patients with Atrial Fibrillation

Minneapolis—Warfarin prophylaxis reduces the rates of primary and secondary stroke in patients with atrial fibrillation (AF), but at a cost—an increased risk of hemorrhage. AF patients who experienced a bleeding event after starting warfarin therapy required more frequent and longer hospitalizations and logged more emergency department (ED) and outpatient visits, according to data presented at the AMCP meeting.

Authors of the Cost of Warfarin-Associated Bleeding in Atrial Fibrillation poster said the increased expense associated with warfarin-triggered bleeding events should be considered when assessing the cost-effectiveness of newer anticoagulant therapies in patients with AF. Researchers reviewed medical and pharmacy claims for individuals in the MarketScan database who had an AF-related claim between January 2003 and December 2007 and identified patients (n=48,069) treated with warfarin within 30 days of the AF claim. They excluded anyone who took warfarin in the 4 months before the AF claim or who had <12 months of follow-up claims. Data were also censored for those whose coverage lapsed or who died during follow-up and for anyone who had multiple types of bleeding events.

At least 95% of individuals who met inclusion criteria were aged ≥50 years in each of the 4 groups: intracranial hemorrhage (ICH), major gastrointestinal (GI) bleed, minor GI bleed, and no bleed. Older patients with more comorbid conditions were most likely to experience a bleeding event (P<.01). In total, 2938 patients suffered a bleeding event within 30 days of their initial warfarin prescription. Most were GI bleeds, with 1811 (3.8%) patients experiencing a minor GI bleed and 926 (1.9%) patients having a major GI bleed. Only 201 (0.4%) patients had an ICH, which is associated with a higher mortality rate than GI bleeds. The mean healthcare expenses per patient in the 12 months after initiating warfarin therapy reached $41,903 in the ICH cohort, $40,586 in the major GI bleed group, and $24,347 in the minor GI bleed cohort. The mean 12- month costs for individuals who did not have a bleed totaled $24,129. Approximately half (50.8%) the costs for patients with an ICH related to the bleeding event compared with 30.8% for patients who had a major GI bleed and 2.6% whose GI bleed was minor.

After adjusting for demographics, insurance status, and comorbidities, data showed annual medical expenses were 61.6% higher in the ICH cohort, 48.7% higher in the major GI cohort, and 0.2% higher in the minor GI cohort compared with the nonbleed (P<.001 for all). The greater number of all-cause hospitalizations among patients with bleed-related claims in the 12-month follow-up period partially accounted for the increased costs. In addition, hospitalizations for patients who experienced a bleed event tended to be longer, with the mean length of stay reaching 13.10 days in the ICH group, 12.88 days in the major GI bleed group, and 7.69 days in the minor GI bleed group versus 7.08 days in the nonbleed cohort. Investigators did not report on ED visits for patients in the nonbleed group, but the mean number of visits for those who did have a bleeding incident ranged from 3.25 to 3.71 days, with more severe bleeds correlating with more trips to the ED. Each of the bleed cohorts also averaged more outpatient visits than the nonbleed group.

Warfarin prophylaxis cuts the risk of stroke in patients with AF by two thirds, said the authors, with an absolute reduction per year of 2.7% for primary stroke and 8.4% for secondary stroke when compared with placebo. This benefit, however, should be weighed against the risk of bleeding events and their costs. The authors suggest that newer agents for anticoagulation therapy might reduce healthcare costs and improve outcomes. This study was supported by Daiichi Sankyo, Inc.