The long-term goal of this project is to develop an effective community-based intervention, using worksites as communities, to increase fruit and vegetable consumption. It builds on the current Seattle 5 a Day project and aims to: 1) Use qualitative methods to modify the community-based 5 a Day intervention for increasing fruit and vegetable consumption to be appropriate for a worker with low socioeconomic status (SES); 2) Develop and automate an additional component, to augment the community-based 5 a Day intervention, which will produce individually tailored information and personalized feedback for low SES workers, 3) Evaluate the effectiveness of this intervention in increasing fruit and vegetable consumption using a rigorous randomized controlled trial of small worksites that employ low SES workers, and 4) Use an objective biomarker (serum carotenoids) to test for intervention-associated response set bias. Variability in cancer mortality rates can be attributed in part to dietary factors, and the Year 2000 goals include a per capita intake of five servings of fruits and vegetables daily. National studies indicate that this goal is not yet attained, and the shortfall in consumption is particularly apparent among adults with low family income. A six-month pilot phase will be used to revise and test the survey instrument, revise the intervention to include personalized tailored feedback, and begin worksite recruitment. In the study phase, 44 worksites will be recruited, all employees will complete baseline surveys, and the worksites will be randomized to intervention or control status. Intervention activities will last 24 months, and will include forming the Employee Advisory Boards, a Kick-off event in each worksite, four personalized tailored feedback mailings to individual employees, and four additional worksite-wide events including a close-out event. The intervention will be followed by a final survey which will occur exactly two years after the baseline. The primary evaluation will use the two cross-sectional surveys, at baseline and follow-up, to assess change in fruit and vegetable consumption, using seven food frequency type summary questions, at the worksite level. Serum carotenoids, as an objective estimator, and data from 24 hour recalls in a subset of respondents will be used to validate results from the main surveys. This rigorous study has the potential to produce an intervention to increase fruit and vegetable intake of demonstrated efficacy, thereby contributing to cancer prevention.