106 Estimation of Medication Adherence by Morisky Medication Adherence Scale-4 Construct Compared to Liquid Chromatography Mass Spectrometry in Emergency Department Patients With Chronic Hypertension

Abstract

Study Objectives: To compare anti-hypertension medication adherence by the subjective MMAS-4 construct with objective LC-MS analysis in emergency department (ED) patients with HTN. According to the 2017 American Heart Association guidelines, 46% of the American population is living with hypertension (HTN). Of the nearly 150 million hypertensive Americans, approximately half do not achieve blood pressure (BP) control due to nonadherence with their prescribed medication treatment. Current measures to evaluate adherence include subjective measures such as the Morisky Medication Adherence Scale-4 (MMAS-4) and direct measures such as quantified medication concentrations using liquid chromatography mass spectrometry (LC-MS). African American patients with chronic HTN enrolled in an ED-based longitudinal study of BP control were treated over 1 year with dispensation and titration of antihypertensive therapy at 9 outpatient follow-up visits using a standardized treatment algorithm. MMAS-4 was scored and serum samples were analyzed using LC-MS at 52 weeks post-randomization to estimate medication adherence. Adherence at 52 weeks by MMAS-4 (Score >1) and LC-MS analysis results (using defined assay quantification cut-points) were compared overall, and among those with and without BP control (defined as systolic < 130mm Hg). 91 patients were included in this analysis (mean age 45.8 +/- 7.9; 50.5% female; mean baseline systolic BP 160.6 +/- 25.8 mmHg; mean systolic BP at week 52 135.1 +/- 18.6 mmHg). Patients were taking an average of 2.4 anti-hypertensive medications. Adherence by MMAS-4 was 85.7% (p=0.607) compared to 61.5% (p=0.127) for the LC-MS analysis overall. Among the 48.4% of patients who achieve BP control, adherence was 83.7% and 69.8% by MMAS-4 and LC-MS; respectively. For 51.6% of patients who did not achieve BP control, adherence was 87.5% and 54.2%; respectively. Medication adherence as determined by the subjective MMAS-4 construct was higher than that obtained by objective LC-MS analysis. This subjective overestimation of adherence did not correlate with BP control highlighting a need for a direct measure of adherence when making clinical decisions that affect medication titration. Furthermore, for patients not achieving BP control, objective analysis demonstrated lower adherence.