Abstract 16454: Medication Adherence and Healthcare Utilization in Heart Failure
Objective: To determine among community patients with heart failure (HF), whether medication adherence is associated with risk of hospitalizations and emergency department (ED) visits, independently of known predictors of outcome.
Background: Medication adherence has been shown to be suboptimal in many HF patients; however, its association with healthcare utilization is less well documented, particularly in the community.
Methods: Olmsted, Dodge and Fillmore County, MN residents with HF were prospectively recruited between October 2007 and November 2009. Pharmacy records were obtained for the 6 months after enrollment and proportion of days covered (PDC) was used to calculate medication adherence for β blockers, ACE/ARBs and statins. A PDC <80% was considered poor adherence. Anderson-Gill modeling, which accounts for repeated events, was used to determine if medication adherence predicted hospitalizations and ED visits.
Results: Among 241 patients (mean age 73±14, 60% male), the proportion of patients prescribed β blockers, ACE/ARBs and statins was 68%, 67% and 59%, respectively. The proportion of poor adherers was 14%, 15% and 10% to β blockers, ACE/ARBs, and statins, respectively. After adjusting for age, sex, ejection fraction and the Charlson comorbidity index, poor adherence to β blockers was associated with an 84% increased risk of hospitalizations and 85% increased risk of ED visits. There was no association between adherence to ACE/ARBs and healthcare utilization; however there was a trend towards an association between poor adherence to statins and hospitalizations and ED visits.
Conclusion: Among community patients with HF, poor adherence to β blockers is a strong and independent predictor of healthcare utilization, underscoring the importance of addressing medication adherence in routine clinical practice.
- © 2011 by American Heart Association, Inc.