摘要:OBJECTIVES. A major assumption underlying youth health promotion has been that physiological risk factors track from childhood into adulthood. However, few studies have systematically examined how behaviors change during adolescence. This paper describes longitudinal tracking of adolescent health behaviors in two Minnesota Heart Health Program communities. METHODS. Beginning in sixth grade (1983), seven annual waves of behavioral measurements were taken from both communities (baseline n = 2376). Self-reported data included smoking behavior, physical activity, and food preferences. RESULTS. A progressive increase in the change to weekly smoking status was observed across the smoking status categories. As students began to experiment with smoking, they were more likely to either begin to be or remain regular smokers. Tracking of physical activity and food choice variables was also apparent. In nearly all the follow-up periods, the students identified at baseline as measuring high remained high, and those measuring low remained low. CONCLUSIONS. These results indicate that there is evidence of early consolidation and tracking of physical activity, food preference, and smoking behavior. The early consolidation of health behaviors implies that interventions should begin prior to sixth grade, before behavioral patterns are resistant to change. The smoking results suggest that students are experiencing difficulty quitting smoking; thus, youth smoking cessation interventions are warranted.