OBJECTIVE: To review the measurement properties (reliability, validity, responsiveness) of the Human Activity Profile (HAP), a self-report measure of energy expenditure or physical fitness. DATA SOURCES: MEDLINE, CINAHL and EMBASE were searched up to September 2005 and the reference lists of included studies were checked for additional relevant studies. REVIEW METHOD: Studies were included that reported Human Activity Profile scores, test-retest reliability, correlations with other measures, or responsiveness (sensitivity to change). Of 83 potentially relevant articles, 39 articles were included plus the test manual. Two independent reviewers extracted data from the included studies. RESULTS: The Human Activity Profile has been used to evaluate physical activity in a wide variety of clinical populations and in healthy individuals. The change in score required to be 90% confident that change is beyond measurement error was estimated to be 7.8 for the Maximum Activity Score and 6.8 for the Adjusted Activity Score. The construct validity of the Human Activity Profile was supported by a large number of studies, although evidence for criterion validity was limited to four studies. No studies have investigated a priori the responsiveness or minimum clinically important difference of the Human Activity Profile. CONCLUSION: The Human Activity Profile appears to be a useful indicator of physical activity levels in people with chronic pain, arthritis, renal failure, various neurological and cardiorespiratory conditions, as well as in healthy older people.