Biological and urban environmental variables as correlates of adolescents' physical activity

The aim of this study was to examine the biological and urban environment variables that associate with physical activity (PA) in adolescents. After this, to examine the interrelationship between biological, urban environment variables and PA in structural equation analysis. This was a cross-sectional study with an adolescents’ randomized sample. Measures included PA (steps per day by pedometers); self-report questionnaire; geographical location; ‘geocoding’ process; and direct observation and cardiorespiratory fitness (by 6-min run test). Linear and binary logistic regression models were tested. In addition, moderation and mediation analysis were tested. The sample consisted of 236 adolescents (61.9% girls) aged 14 to 18 years. The commuting to school was associated with residence distance to school (OR = 6.41; CI95%: 1.01-40.80) and walkability (OR = 1.40; CI95%: 1.02-1.94). The gender moderates the relationship between walkability and commuting to school, association only in girls (OR = 1.72; p < 0.05). The relationship between the use of public spaces and PA was reduced (Δβ = -1320.6 steps/day; p < 0.05) in the presence of cardiorespiratory fitness (mediation effect). In conclusion, adolescents’ PA is associated with the use of public spaces, but this relation is mediated by cardiorespiratory fitness. In addition, PA is associated with commuting to school. This commuting is associated with residence distance to school and walkability just in girls.