Objectives: The main objective of this paper is to describe the approach and specific findings of the European Physical Activity Surveillance System (EUPASS) research project. In particular, the analysis presented aims at testing the reliability, comparability and predictive power of different sets of physical activity (PA) indicators. Design: First, a panel study based on computer-aided telephone interview (CATI) was designed to report PA data of a representative, selected group of about 100 persons per country at three points in time. Second, a CATI time series survey was carried out with the goal of realising about 100 interviews per month over six consecutive months. Setting: The project was carried out in eight European countries to support the development of the European Union's (EU) Health Monitoring Programme. Subjects: Random population samples (subjects aged 18 years and older) were drawn from each participating country. Results: While many PA indicators used in EU countries to date as well as the psychosocial and environmental measures tested in the present study had acceptable to good reliability coefficients, the test–retest reliability scores of the International Physical Activity Questionnaire (IPAQ) version tested (the short (last 7 days) telephone interview IPAQ; IPAQ-S7T) were rather low. The comparability between extant national PA items and the IPAQ-S7T was low for all countries. The strongest predictors of perceived health were the psychosocial and environmental PA indicators. Conclusions: According to the results of the present study, more research is needed to further investigate and improve the quality of the IPAQ. In addition, the specific predictive power of the tested psychosocial and environmental PA indicators on perceived health should be of particular interest for designing health surveillance activities in the future.