BACKGROUND: Research consistently demonstrates that physical activity declines with age. However, such declines do not occur linearly. The transition into early adulthood is one period in which disproportionate declines in physical activity have been evident, but much of our understanding of such declines among young adults has been based on either cross-sectional data or prospective studies that focus exclusively on college/university students. PURPOSE: The purpose of the current study was to use multilevel modeling to discern patterns of physical activity based on gender and educational trajectory among a nationally representative cohort of Canadian adolescents (N=640; ages at baseline, 12-15 years). Examinations of smoking and binge drinking also were conducted as a basis for comparison. METHODS: Drawn from seven cycles of the National Population Health Survey (NPHS), participants were interviewed every 2 years from 1994-1995 to 2006-2007; data analysis was conducted in 2010. RESULTS: Overall, there was a 24% decrease in physical activity (equivalent to 1 MET/day) across the 12-year period. A significant three-way time X gender X educational trajectory interaction (coefficient= -0.189, SE=0 .09, p<0.05) emerged in the physical activity analysis; main effects in time(2) (coefficient= -0.114, SE=0.01, p<0.01) and time(3) (coefficient=0.028, SE=0.01, p<0.01) were significant for binge drinking and smoking, respectively. CONCLUSIONS: Physical activity decline was evident during young adults' transition into early adulthood, with declines being steepest among men who entered a college/university. Although there were increases in several health-risk behaviors during adolescence, individuals tend to grow out of binge drinking and smoking as they mature.
BACKGROUND: Research consistently demonstrates that physical activity declines with age. However, such declines do not occur linearly. The transition into early adulthood is one period in which disproportionate declines in physical activity have been evident, but much of our understanding of such declines among young adults has been based on either cross-sectional data or prospective studies that focus exclusively on college/university students. PURPOSE: The purpose of the current study was to use multilevel modeling to discern patterns of physical activity based on gender and educational trajectory among a nationally representative cohort of Canadian adolescents (N=640; ages at baseline, 12-15 years). Examinations of smoking and binge drinking also were conducted as a basis for comparison. METHODS: Drawn from seven cycles of the National Population Health Survey (NPHS), participants were interviewed every 2 years from 1994-1995 to 2006-2007; data analysis was conducted in 2010. RESULTS: Overall, there was a 24% decrease in physical activity (equivalent to 1 MET/day) across the 12-year period. A significant three-way time X gender X educational trajectory interaction (coefficient= -0.189, SE=0 .09, p<0.05) emerged in the physical activity analysis; main effects in time(2) (coefficient= -0.114, SE=0.01, p<0.01) and time(3) (coefficient=0.028, SE=0.01, p<0.01) were significant for binge drinking and smoking, respectively. CONCLUSIONS: Physical activity decline was evident during young adults' transition into early adulthood, with declines being steepest among men who entered a college/university. Although there were increases in several health-risk behaviors during adolescence, individuals tend to grow out of binge drinking and smoking as they mature.