Medication Adherence to Specific Drug Classes in Chronic Heart Failure

BACKGROUND: Adherence to medication is crucial to improve clinical outcomes in patients with heart failure (HF). However, at least 1 out of 4 patients is nonadherent to his or her medication. Several studies have quantified medication adherence in HF patients, monitoring only 1 drug with the Medication Event Monitoring System (MEMS). Some authors have argued that monitoring 1 drug reflects the whole adherence behavior, although there is some evidence of important differences in adherence to distinct drug classes. Furthermore, medication characteristics could be a relevant predictor of adherence, and different drugs could pose different barriers to patients. OBJECTIVES: To (a) quantify medication adherence to angiotensin-converting enzyme inhibitors (ACEI), beta blockers, and loop diuretics and (b) compare the agreement in adherence among drug classes in chronic HF. METHODS: Medication adherence to 3 different drugs was monitored using MEMS in 63 patients (81% male, mean age 63.5 years). Medication adherence was measured as the percentage of prescribed doses effectively taken. Patients were considered to be adherent when at least 88% of prescribed doses were taken. Adherence agreement between drug classes was analyzed with Bland-Altman plots and Kappa coefficients. RESULTS: The mean adherence was 97.3% for ACEI, 97.2% for beta blockers, and 96.0% for loop diuretics. Individual patients did not adhere equally to all drug classes, with differences within the same patient ranging from -35% to 33%. The proportion of patients classified as adherent was 77.8% to ACEI, 69.8% to beta blockers, and 69.8% to loop diuretics. The agreement between each of 2 drugs regarding adherence was substantial (beta blocker vs. ACEI: K = 0.72; beta blocker vs. diuretic: K = 0.62; ACEI vs. diuretic: K = 0.72). If patients were classified as adherent and nonadherent based only on 1 drug, 20% of patients would be misclassified regarding the other drugs. CONCLUSIONS: Patients can adhere differently to medication used in HF treatment, with lowest adherence to loop diuretic and beta blockers and highest adherence to ACEI. Studies measuring medication adherence should always specify the drug class being analyzed and should not mix different drug classes to generalize about adherence behavior.