Changes in and the mediating role of physical activity in relation to active school transport, fitness and adiposity among Spanish youth: the UP&DOWN longitudinal study

BACKGROUND: Longitudinal changes in child and adolescent active school transport (AST), and the mediating role of different intensities of daily physical activity (PA) levels in relation to AST and physical fitness and adiposity indicators is unclear. This study aimed to: 1) describe longitudinal changes in AST, light PA (LPA), moderate- to vigorous-intensity PA (MVPA), physical fitness and adiposity indicators over three time-points; and 2) investigate the mediating role of LPA and MVPA levels on associations between AST and physical fitness and adiposity indicators over three time-points among children and adolescents. METHODS: This longitudinal study comprised 1646 Spanish children and adolescents (48.8% girls, mean age 12.5 years ±2.5) at baseline, recruited from schools in Cádiz and Madrid. Mode of commuting to school was self-reported at baseline (T0, 2011–12), 1-year (T1) and 2-year follow-up (T2). PA was assessed using accelerometers. Handgrip strength, standing long jump and cardiorespiratory fitness (CRF) assessed physical fitness. Height, weight, body mass index, waist circumference, and triceps and subscapular skinfold thickness were measured. Multilevel linear regression analyses assessed changes in AST, PA levels, physical fitness and adiposity indicators over three time-points (T0-T1-T2). Additionally, longitudinal path analysis (n = 453; mean age [years] 12.6 ± 2.4) was used to test the mediating effects of LPA and MVPA levels on the association between AST and physical fitness and adiposity indicators. RESULTS: Multilevel analyses observed decreases in LPA between T0-T1 (β = − 11.27; p < 0.001) and T0-T2 (β = − 16.27; p < 0.001) and decreases in MVPA between T0-T2 (β = − 4.51; p = 0.011). Moreover, changes over time showed increases in handgrip between T0-T1 (β = 0.78; p = 0.028) and T0-T2 (β = 0.81; p = 0.046). Path analyses showed that AST was directly positively associated with MVPA at T1 (all, β ≈ 0.33; p < 0.001). MVPA at T1 mediated associations between AST and CRF at T2 (β = 0.20; p = 0.040), but not the other outcomes. LPA did not mediate any associations. CONCLUSIONS: Results from longitudinal path analysis suggest that participation in more AST may help attenuate declines in MVPA that typically occur with age and improve CRF. Therefore, we encourage health authorities to promote AST, as a way to increase MVPA levels and CRF among youth.