Psychometric Properties of a Short Self-Reported Measure of Medication Adherence Among Patients With Hypertension Treated in a Busy Clinical Setting in Korea

BACKGROUND We examined the psychometric properties of the Korean version of the 8-item Morisky Medication Adherence Scale (MMAS-8) among adults with hypertension. METHODS A total of 373 adults with hypertension were given face-to-face interviews in 2 cardiology clinics at 2 large teaching hospitals in Seoul, South Korea. Blood pressure was measured twice, and medical records were reviewed. About one-third of the participants (n = 109) were randomly selected for a 2-week test-retest evaluation of reliability via telephone interview. RESULTS Internal consistency reliability was moderate (Cronbach α = 0.56), and test-retest reliability was excellent (intraclass correlation = 0.91; P < 0.001), although a ceiling effect was detected. The correlation of MMAS-8 scores with scores for the original 4-item scale indicated that convergent validity was good (r = 0.92; P < 0.01). A low MMAS-8 score was significantly associated with poor blood pressure control (χ(2) = 29.86; P < 0.001; adjusted odds ratio = 5.08; 95% CI, 2.56-10.08). Using a cut-off point of 6, sensitivity and specificity were 64.3% and 72.9%, respectively. Exploratory factor analysis identified 3 dimensions of the scale, with poor fit for the 1-dimensional construct using confirmatory factory analysis. CONCLUSIONS The MMAS-8 had satisfactory reliability and validity and thus might be suitable for assessment and counseling regarding medication adherence among adults with hypertension in a busy clinical setting in Korea.