Abstract 148: Association of Dosing Frequency on Medication Adherence in Patients with Non-Valvular Atrial Fibrillation

Abstract

Objective: Previous studies suggest that the complexity of a dosing regimen may affect medication adherence. We examined the association between dosing frequency and adherence for 2 concomitant medications commonly prescribed to patients with non-valvular atrial fibrillation (NVAF), metoprolol (MET) and carvedilol (CAR). Methods: A retrospective claims study from a large US commercial and Medicare Advantage health plan analyzed data of adults ( > 18 years) with 1 inpatient or 2 outpatient claims for NVAF between 1/1/2008 - 12/31/2010. Patients with > 2 pharmacy claims for MET or CAR were analyzed separately. Within MET and CAR samples, once-daily (QD) and twice-daily (BID) cohorts were defined by the dosing frequency on pharmacy claims. The index date was set as the date of the first MET or CAR claim. Patients were continuously enrolled in the health plan for 1 year before (pre-index) and 1 year after (post-index) the index date. MET patients were required to have > 1 pre-index claim for acute myocardial infarction, angina, heart failure, or hypertension; CAR patients were required to have ≥1 claim for heart failure or hypertension. Patients using both QD and BID formulations of the index medication were excluded. Adherence to the index medication was assessed by the proportion of days covered (PDC) during the post-index period. PDC between QD and BID patients was compared using logistic regression to adjust for demographic and pre-index clinical characteristics. The proportion of QD and BID patients who discontinued the index medication (defined by a gap > 30 days) during the post-index period was also compared. Results: The analysis included 11,621 MET patients (QD: 6,084; BID: 5,537) and 4,393 CAR patients (QD: 203; BID: 4,190). Mean (SD) age was 70 (12) years for MET and CAR patients; 59% of MET and 69% of CAR patients were male. Compared to patients with BID dosing, patients with QD dosing were on average younger, more likely to be male, and had a lower comorbidity burden. Fewer patients discontinued MET or CAR with QD than BID dosing (MET: 38% vs. 51%, p<0.001; CAR: 39% vs. 48%, p=0.009). The proportion of patients with PDC > 80% was greater for patients with QD than BID dosing (MET: 62% vs. 50%, p< 0.001; CAR: 63% vs. 53%, p=0.004). MET patients with BID dosing were less likely to achieve PDC > 80% than patients with QD dosing (adjusted OR: 0.66; 95% CI: 0.609-0.712). CAR patients with BID dosing were less likely to achieve PDC > 80% than patients with QD dosing (adjusted OR: 0.69; 95% CI: 0.508-0.934). Among MET and CAR patients, age <60 years was associated with lower adherence (p<0.001) while prior use of index medication was associated with higher adherence (p≤0.001) to the index medication. Conclusion: Medication adherence to MET and CAR was higher with QD than BID dosing. Quality initiatives that reduce the dosing frequency of treatment regimens may improve medication adherence among NVAF patients.