Validity and reliability of a self-reported measure of medication adherence in patients with Type 2 diabetes mellitus in Singapore

• Published in: Diabetic medicine 2012-09, Vol.29 (9), p.e338-e344
• Main Authors: Wang, Y., Lee, J., Toh, M. P. H. S., Tang, W. E., Ko, Y.
• Format: Article
• Language: English
• Subjects: Aged; Cross-Sectional Studies; diabetes mellitus; Diabetes Mellitus, Type 2 - drug therapy; Diabetes Mellitus, Type 2 - epidemiology; Diabetes Mellitus, Type 2 - ethnology; Female; Humans; Hypoglycemic Agents - therapeutic use; Male; medication adherence; Medication Adherence - psychology; Medication Adherence - statistics & numerical data; Middle Aged; Morisky-Green-Levine Medication Adherence Scale; Outpatients; Patient Compliance - psychology; Patient Compliance - statistics & numerical data; Psychometrics; Reproducibility of Results; Retrospective Studies; Self Report; Singapore - epidemiology
• ISSN: 0742-3071; 1464-5491
• DOI: 10.1111/j.1464-5491.2012.03733.x

Diabet. Med. 29, e338–e344 (2012) Aims  A reliable and valid measure is essential for the assessment of medication adherence. Until now, no patient‐reported medication adherence measure has been validated in Singapore. The aim of this study was to validate a modified 4‐item Morisky–Green–Levine Medication Adherence Scale in patients with Type 2 diabetes in Singapore. Methods  A cross‐sectional survey was conducted in a sample of outpatients with Type 2 diabetes in Singapore from September to December in 2009. Respondents completed either an English or Chinese version of the modified 4‐item Morisky–Green–Levine Medication Adherence Scale. The scale scores ranged from 0 to 4, with higher scores indicating better medication adherence. Reliability was assessed using Cronbach’s alpha. Content validity was assessed by expert review. Construct validity was examined using factor analysis and hypothesis testing. Results  Of the 294 respondents who completed the modified Morisky–Green–Levine Medication Adherence Scale, 13.3, 21.4, 35.7 and 29.6% had a score of 0–1, 2, 3 and 4, respectively. The internal consistency of the scale was moderate (Cronbach’s alpha = 0.62). Principal component analysis showed that the four items loaded onto one factor (eigenvalue = 1.95). Respondents with higher scores were older (P