PCV108 Deriving Utility Values from the General Population for Dronedarone in the Treatment of Atrial Fibrillation

Abstract

OBJECTIVES:In this study, we seek to measure the association between physicianrated adherence to antihypertensive treatment and blood pressure outcomes. Prior to the study, we hypothesized that systolic and diastolic blood pressure outcomes were also related with patient and physician variables. Therefore, we also seek to determine the impact of these variables on the outcome. METHODS: Unlike prior adherence studies, which have consulted small sample sizes by means of timeconsuming and complex questionnaires, our research pools the findings from six Valsartan studies, thereby evaluating 15,583 patients, while utilizing a simple and cost-effective survey method. Adherence was measured using a four-item yes/no Basel Assessment for Adherence Scale (BAAS), in which a "yes" answer on any item classified a patient as non-adherent. The measurements were performed at baseline and at 90 days. Hierarchical logistic regression was used to identify patientand physician-related determinants. RESULTS: Our results indicate that inadequate blood pressure control correlates with poor adherence at baseline and 90 days. The mean absolute difference in blood pressure between those who were adherent (70% of the patient population) and those who were not was 5 to 19 mmHg (pϽ0.001). Every adherent patient was likely to decrease his systolic and diastolic blood pressure by 2.3 and 1.3 mmHg, respectively (pϽ0.001). The majority of the variability in blood pressure values, however, was due to patient variables, many of which can be managed with medical care, or identified in patients with higher risks, poor adherence, or poly-pharmacy. CONCLUSIONS: Physician-reported adherence by means of a short patient BAAS questionnaire is a direct, simple and especially inexpensive method to assess adherence status. This method can be easily integrated into routine clinical practice and provides evidence that adherence is positively correlated with blood pressure reduction in hypertensive patients. However, patient-related variables also correlate strongly with the outcome and demand further investigation.