The purpose of this study is to test the hypothesis that increased involvement in an exercise program deters smoking initiation, especially among high risk youth who have not responded to traditional smoking prevention techniques. In order to obtain the health objective of reduced smoking initiation, three specific aims are proposed: (a) to test the effectiveness of an exercise intervention to prevent smoking initiation, (b) to examine the relative effectiveness of the exercise intervention for students identified as especially high risk, and (c) to investigate mechanisms by which exercise exerts its influence in the prevention of smoking. All subjects in the study will be high risk, based on their low socioeconomic status. Approximately 3500 students from three school districts will complete a baseline survey at the end of elementary school. The school districts were identified according to four criteria: sufficient size of the district, low socioeconomic status, ethnic diversity, and proximity to the research center. The students will be followed into four middle schools per district. In each case the four schools will be matched, forming two pairs; one member of each pair will be randomly assigned to the exercise intervention and the other will be assigned to the no treatment control. These cohorts will receive fitness testing pre and post intervention and will maintain exercise logs. The intervention consists of an on-going exercise program that is comprised of traditional sports, recreational aerobic activities, and opportunities for lifestyle skill development. It will be administered three days per week for twenty weeks of each school year. First, the efficacy of the intervention on preventing smoking initiation will be evaluated both for the students in general and for a subset of subjects identified as """"""""high/high risk."""""""" Second, using Jessor and Jessor's Problem Behavior Theory, the effect of various baseline psychosocial variables on smoking initiation and the interaction of these characteristics with the exercise intervention will be assessed to explore the mechanisms by which exercise might yield its effect. To analyze the data, a hierarchical logistic regression will be employed with the log odds ratio of smoking status serving as the dependent variable. Findings in the predicted direction would have implications for prevention as there would be evidence that this relatively low-cost intervention, an exercise program, would be useful in reaching a heretofore difficult-to-reach population and deter smoking initiation.
