Development and psychometric testing of the medication taking behavior tool in Thai patients

The previous Thai version of the 8-item Morisky Medication Adherence Scale (8-item MMAS) showed poor sensitivity and unacceptable internal consistency reliability.

To develop and test the psychometric properties of a new medication taking behavior measure for Thai patients (MTB-Thai) including practicality, reliability and validity.

This study was conducted with adult outpatients regularly taking any medicines for at least three months from three university hospitals in Bangkok, Thailand, between July 2014 and March 2015. The study was approved by the Ethical Committees of the three hospitals. Practicality was assessed by administration time and the percentage of missing data. Internal consistency and test-retest reliability were evaluated employing Cronbach's alpha and intraclass correlation coefficients (ICCs), respectively. Validity was evaluated with content, construct, convergent and known-groups validity.

Of 1156 patients, the 6-item MTB-Thai had an average administration time of 2 min and no missing data. It showed good Cronbach's alpha value of 0.76 and excellent ICCs of 0.83. The MTB-Thai showed good content validity with the item level and scale level of content validity indexes greater than the acceptable levels of 0.8 and 0.9, respectively. Both exploratory and confirmatory factor analyses showed the MTB-Thai had two domains including unintentional and intentional domains. The MTB-Thai correlated well with the overall medication adherence scale with a Pearson's correlation coefficient of 0.62 (p < 0.001). The sensitivity, specificity, positive predictive value and negative predictive value of the MTB-Thai were 76, 35, 55 and 57 %, respectively.

The MTB-Thai was practical, reliable and valid in assessing mediation taking behaviors in Thai patients with chronic diseases.