SUBJECTS AND DESIGN: The study was based on the total patient sample (n = 110) of a randomized controlled trial comparing two intervention methods advising cardiovascular high-risk men of lifestyle changes in general practice. Behaviour and risk factor changes during the one-year intervention study were analysed using multiple regression and logistic regression analyses with the above-mentioned independent variables. SETTING: Twenty-two general practice centres in the county of Hordaland, western Norway. RESULTS: Self-efficacy of increased physical exercise was the only variable significantly related to exercise change. Age and self-efficacy were statistical significant predictors of smoking cessation success. None of the independent variables was statistically significantly related to blood pressure or cholesterol change. Educational level related negatively, although statistically insignificantly, with total risk change. CONCLUSION: The study confirms the importance of self-efficacy in both human behaviour and motivation for behaviour change. OBJECTIVES: The objectives of the study were to explore the impact of possible predictors for cardiovascular risk behaviour change, predictors such as education, age, self-efficacy, doctors' interpersonal skills, and number of appointments.