Objective
 In recent years, antiretroviral therapy (ART) has been significantly expanded in developing countries, while drug resistance to HIV caused by low adherence is becoming a grave concern. As a member of the international community, Japan is expected to expand its cooperation for supporting the expansion of ART. However, the evaluation of ART adherence remains a challenge since the definition and the methods of its measurement are not standardized. In this regard, the articles of studies on ART adherence are reviewed to investigate available methodologies that can be used for measurement.
Method
 Articles were searched and extracted through Ovid Full Text database for the period between Jan. 2002 and Aug. 2006 by using keywords of “adherence” and “HIV”. Among 81 extracted original articles, 50 articles were selected based on the inventory and clear identification of the methodologies used to measure adherence.
Result
 The studies were conducted in the US (28 articles: 56%), Canada (5 articles: 10%), UK (3 articles: 6%), Africa and South America (10 articles: 20%) and no articles were extracted from Asia. The mean sample size of the studies was 581.2 (range: 24-6288). Measurements of adherence that were used in the articles as follows; patient's self-report (31 articles: 62%), electric drug monitoring (14 articles: 28%), pharmacy's refill record (12 articles: 24%), pill-count (9 articles: 18%), laboratory testing (6 articles: 12%) and combination of these (14 articles: 28%). Of the 31 articles using patient's self-report, 25 articles asked for the participant's frequency of missed dose.
Conclusion
 Studies concerned with ART adherence have been mainly undertaken in industrialized countries, and it was found that inquiries on missed doses were the most frequently used method to measure ART adherence. We strongly suggest the development of more simplified methods for measuring ART adherence, especially for resource-limited settings.