This study will test a 24-month multi-component intervention designed to increase physical activity, reduce sedentary practices and promote healthy diets among preteens who obtain orthodontia care. The intervention is based on the Behavioral Ecological Model (BEM), where interacting physiological, environmental, and cultural contingencies select behavior. This study will require five years to recruit 50 orthodontist offices, and 2100 youth, and implement an office-wide intervention for promoting healthy diet and physical activity among 8-13 yr old youth. Our primary aims are to detect a differential change in diet, physical activity and BMI in our experimental group compared to the tobacco prevention control group; to observe differential effects of the intervention between low vs. high SES; and similarly for males vs. females; to validate our primary outcome variables by contrasting them with a "gold standard"; and determine the degree to which family and peer encouragement for sedentary and high caloric diet at baseline increases the relative effect of the orthodontist intervention. Offices will be assigned at random to condition producing a 2 (experimental group) by 2 strata (low vs. high SES) by 2 strata (male vs. female) by 4 (repeated measures) experimental design. Those in the experimental condition will be exposed to the office-wide intervention for diet and activity; those in the control condition will be exposed to an office-wide intervention for tobacco prevention. The intervention will consist of training of orthodontists and office staff to provide appropriate social prompts, reinforcement and counseling, and modification of the offices to provide supportive physical and social resources for their patients and parents. Measures of diet, physical activity, body composition, tobacco use, fitness and family and peer encouragement will occur at baseline, 12, 24, and 36 months. Differential change over time will be analyzed using Random Effects Multivariate modeling. Analyses will include control for office cluster effects and multivariate models will include analyses of hypothetical main effects, experimental group by time interactions, and moderating effects by inclusion of interaction terms for susceptibility by experimental condition and other possible moderators. Relevance: If successful, this study may inform policies that would promote all specialists (physicians, general dentists) to deliver brief counseling for diet and physical activity. Doing so might yield the cumulative exposure necessary to effect and sustain change in a large proportion of the preteen population.