Study characteristics
The characteristics of the studies included in the research are tabulated in Table 1. The published articles were of worldwide origins and were conducted in countries including the United Kingdom, Italy, the United States, Canada, Australia, Singapore, Malaysia, Thailand, Iran, Egypt, Syria, and India. Most of the studies were cross-sectional studies (N=16, 80% ) [11-26], followed by cohort studies (N=3, 15%) [27-29], and only one randomised controlled trial [30]. The majority of the studies were single-centred (N=13, 65%) and about 30% of the studies were conducted in multiple settings [12, 17, 24, 25, 27, 30], while the remaining study did not mention the setting [11]
A variety of medication measures was used across the studies. Most of the studies (N= 7, 35%) examined the rate of compliance using the frequency of ICT’s administration [11, 12, 14, 16, 18, 19, 23], followed by studies that evaluated patients’ compliance status using vial or pill count (N=5, 25%) [13, 27-30]. Several studies (N = 3, 15%) used the Likert Scale [17, 21, 26], while some studies (N = 3, 15%) measured compliance using self-reported questionnaires such as Morisky Medication Adherence Scale (MMAS-8) [22], the Medication Compliance Questionnaire (MCQ) [24], and standardised questionnaire (not mentioned in the study) [25]. Meanwhile, a study [20] reported the compliance with medication possession ratio (MPR), and another study [15] did not mention the method of compliance measurement.