Individual, family, school and neighborhood predictors related to different levels of physical activity in adolescents: a cross-sectional study.
Description
The aim of this study was to investigate the association among individual, family, school environment and neighborhood predictors with the different levels of physical activity (PA) [light (LPA) and moderate to vigorous PA (MVPA)] in Brazilian adolescents. A cross-sectional study was carried out with 309 adolescents with a mean age of 15.37 (± 0.57) years. PA and sleep time were assessed by accelerometry. Individual predictors were determined by anthropometry and questionnaires, while family, school environment and neighborhood predictors were assessed using questionnaires. Robust Regression analysis was performed considering a significance level of 5%. Individual and environmental variables were able to respectively predict 64% and 13.6% of adolescents’ participation in LPA. Work (ꞵp = 0.2322), gender (ꞵp = -0.1318), commuting to school (ꞵp = -0.1501), sleep (ꞵp = -0.1260) and paved roads (ꞵp = -0.1360) were associated with LPA. It was also observed that individual (59.4%) and environmental (27.4%) variables were able to predict adolescents’ participation in MVPA. Work (ꞵp = 0.1656), commuting to school (ꞵp = 0.1242) and crime (ꞵp = 0.1376, and gender (ꞵp = - 0.3041) and paved roads (ꞵp = -0.1357 were associated with MVPA. Such results indicated that boys, those who work and those who live in unpaved neighborhoods presented greater time in LPA and MVPA; those who live in neighborhoods with higher crime had higher time spent in MVPA; and those who passively commute to school had more time in LPA. There was an average reduction of 5.0 minutes in LPA time for each additional hour of sleep. Finally, students who actively commute to school had more time in MVPA. Individual factors and those related to the neighborhood environment can play an important role in understanding the variables which can influence the different levels of PA in adolescents.