OBJECTIVES As electronic versions of HRQL measures such as the EQ-5D become available, it is important to understand the reliability of different modes of technology. The aim of this study was to summarize the evidence of test-retest reliability for the EQ-5D visual analog scale (VAS), a scale ranging from 0 (worst imaginable health) to 100 (best imaginable health). METHODS A structured literature search was conducted in MEDLINE using keywords relevant to EQ-5D, visual analog scales, and test-retest reliability. Original research studies that reported information on the test-retest reliability of the EQ-5D VAS were included. Demographic characteristics, interval between observations, and intraclass correlation coefficients (ICCs) were abstracted. RESULTS Of the 25 studies that examined test-retest reliability of EQ-5D, 14 reported evidence of test-retest reliability for EQ-5D VAS. Most of the papers were studies that assessed the validity of EQ-5D for certain countries or languages (n= 5/14, 36%) or for use in patient groups / certain medical conditions (n= 7/14, 50%). The most common interval between observations was 2 weeks (n=4/14, 29%), with analyses conducted on a subgroup of self-reported stable patients, based on self-report, in 4 studies. TRT ICCs ranged from ICC=0.38 for Alzheimer's patients to ICC=0.90 (95% CI: 0.88-0.92) for a methodological study conducted in Spain, with a median ICC=0.8 across the 14 studies. Almost 80% of studies (11/14) reported ICCs above 0.7, a reliability threshold considered acceptable at the group level. CONCLUSIONS EQ-5D VAS demonstrated acceptable TRT reliability in most studies of populations and medical conditions except in Alzheimer's disease, where proxies but not patients provided reproducible assessments.