Abstract 16864: Patterns of Adherence to Dabigatran and its Association With Outcomes
Background: Dabigatran is a novel oral anti-coagulant that reduces the risk of stroke in atrial fibrillation (AF) patients. Since routine monitoring with laboratory tests is not required, concerns have been raised about the potential for non-adherence to dabigatran. To date, neither the prevalence of non-adherence to dabigatran in routine clinical practice nor the association between non-adherence and clinical outcomes are known.
Methods: We studied a national cohort of AF patients (n=6,256) who filled at least 1 new dabigatran prescription between December 1, 2010 and December 31, 2012 based on Veterans Affairs pharmacy records. Adherence in the year after starting dabigatran was evaluated using proportion of days covered (PDC; number of days supplied divided by observation period). Then, we assessed the association between adherence and adverse outcomes (composite of stroke and all-cause mortality) using Cox Proportional Hazards regression with adherence as a time-varying covariate and adjusting for demographic and clinical variables.
Results: The average age was 71 years, 93% were males and comorbidities were common, including CHF (40%), diabetes (45%), hypertension (93%) and prior stroke (21%). Median PDC in the year after starting dabigatran was 0.66 (IQR 0.33 - 0.90; Figure). Over three-quarters (81%) of patients had a gap in therapy with a median of 2 gap days. There were 5.5% death/stroke events over a median follow-up of 294 days. In multivariable analysis, lower adherence to dabigatran was associated with higher risk of death/stroke (HR 1.20 per 0.1 unit lower PDC, 95% CI 1.12 - 1.27). Findings were consistent for risk of stroke only (HR 1.13 per 0.1 unit lower PDC, 95% CI 0.98 - 1.28).
Conclusion: In the year after starting therapy, dabigatran adherence was suboptimal. Further, lower levels of adherence were associated with adverse outcomes. These findings suggest that efforts to improve adherence to dabigatran are needed in routine clinical practice.
- © 2013 by American Heart Association, Inc.