BACKGROUND: Published research on the use of Web-based behavior change programs is growing rapidly. One of the observations characterized as problematic in these studies is that participants often make relatively few website visits and spend only a brief time accessing the program. Properly structured websites permit the unobtrusive measurement of the ways in which participants access (are exposed to) program content. Research on participant exposure to Web-based programs is not merely of interest to technologists, but represents an important opportunity to better understand the broader theme of program engagement and to guide the development of more effective interventions. OBJECTIVES: The current paper seeks to provide working definitions and describe initial patterns of various measures of participant exposure to ChewFree.com, a large randomized controlled trial of a Web-based program for smokeless tobacco cessation. METHODS: We examined measures of participant exposure to either an Enhanced condition Web-based program (interactive, tailored, and rich-media program) or a Basic condition control website (static, text-based material). Specific measures focused on email prompting, participant visits (number, duration, and pattern of use over time), and Web page viewing (number of views, types of pages viewed, and Web forum postings). RESULTS: Participants in the ChewFree.com Enhanced condition made more visits and spent more time accessing their assigned website than did participants assigned to the Basic condition website. In addition, exposure data demonstrated that Basic condition users thoroughly accessed program content, indicating that the condition provided a meaningful, face-valid control to the Enhanced condition. CONCLUSIONS: We recommend that researchers conducting evaluations of Web-based interventions consider the collection and analysis of exposure measures in the broader context of program engagement in order to assess whether participants obtain sufficient exposure to relevant program content. [J Med Internet Res 2006;8(3):e15]