Background: In recent years, many tailored lifestyle counseling programs have become available through the Internet. Previous research into such programs has shown selective enrollment of relatively healthy people. However, because of the known dose-response relationship between the intensity and frequency of counseling and the behavior change outcomes, selective retention may also be a concern. Objective: The aim of this study was to identify rates and determinants of repeat participation in a Web-based health behavior change program. Methods: A Web-based health behavior change program aimed to increase people's awareness of their own lifestyle, to promote physical activity, and to prevent overweight and obesity was available on the Internet from July 2004 onward at no cost. Univariate and multivariate logistic regression analyses were conducted to identify characteristics of people who participated in the program more than once. Age, compliance with physical activity guidelines, body mass index, smoking status, and the consumption of fruit, vegetables, and alcohol were included in the analyses. Results: A total of 9774 people participated in the baseline test, of which 940 used the site more than once (9.6%). After exclusion of individuals with incomplete data, 6272 persons were included in the analyses. Of these 6272 people, 5560 completed only the baseline test and 712 also participated in follow-up. Logistic regression predicting repeated use determined that older individuals were more likely to participate in follow-up than people aged 15-20 years. The odds ratios for the age categories 41-50, 51-60, and > 60 years were 1.40 (95% CI = 1.02-1.91), 1.43 (95% CI = 1.02-2.01), and 1.68 (95% CI = 1.03-2.72), respectively. Individuals who never smoked were more likely to participate repeatedly than current smokers and ex-smokers (OR = 1.44, 95% CI = 1.14-1.82 and OR = 1.49, 95% CI = 1.17-1.89, respectively). People meeting the guidelines for physical activity of moderate intensity (OR = 1.23 95% CI = 1.04-1.46) and for vegetable consumption (OR = 1.26 95% CI = 1.01-1.57) were also more likely to participate repeatedly than people who did not, as were obese people compared to individuals with normal weight (OR = 1.41 95% CI = 1.09-1.82). Conclusions: For some variables, this study confirms our concern that behavioral intervention programs may reach those who need them the least. However, contrary to most expectations, we found that obese people were more likely to participate in follow-up than people of normal body weight. The non-stigmatizing way of addressing body weight through the Internet may be part of the explanation for this. Our findings suggest that Web-based health behavior change programs may be more successful in the area of weight management than in many other health-related areas. They also stress the importance of adequate coverage of weight management in Web-based health promotion programs, as a driver to continue participation for overweight and obese people.