A B S T R A C TPurpose: Low levels of physical activity (PA) have been shown to be associated with depression in adults. The few studies that focused on adolescents yielded mixed and inconsistent results. Efforts to examine the direction of this relationship have been inconclusive up to now. The aims of this study were therefore to investigate (1) the direction of the inverse association between PA and depressive symptoms over time, and (2) whether these associations are specific to particular clusters of depressive symptoms in adolescents. Methods: Depressive symptoms and PA were assessed in a population sample of adolescents (N ϭ 2,230) who were measured at three waves between age 10 and age 17. Depressive symptoms were measured by the Affective Problems scale of the Youth Self-Report and Child Behavior Checklist, whereas PA was operationalized as the amount of time spent on physical exercise. Structural equation modeling was used to examine bidirectional effects of PA and depressive symptoms over time. Results: We found significant cross-lagged paths from prior PA to later depression as well as from prior depression to later PA (beta values ϭ Ϫ.039 to Ϫ.047). After subdividing depression into affective and somatic symptoms, the affective symptoms were reciprocally related to PA, whereas the paths between somatic symptoms and PA did not reach statistical significance. Conclusions: An inverse bidirectional association between PA and general depressive symptoms was observed. This association was restricted to affective symptoms.

This study showed that the association between physical activity and affect varied considerably across individuals. Thus, while at the group level the effect of physical activity on affect may be small, in some individuals the effect may be clinically relevant.

Objective: Physical activity is inversely associated with depression in adolescents, but the overall associations are fairly weak, suggesting individual differences in the strength of the associations. The aim of this study was to investigate whether plasticity genes modify the reciprocal prospective associations between physical activity and depressive symptoms found previously. Methods: In a prospective population-based study (N = 1,196), physical activity and depressive symptoms were assessed three times, around the ages of 11, 13.5, and 16. Structural Equation Modeling was used to examine reciprocal effects of physical activity and depressive symptoms over time. The plasticity genes examined were 5-HTTLPR, DRD2, DRD4, MAOA, TPH1, 5-HTR2A, COMT, and BDNF. A cumulative gene plasticity index consisting of three groups (low, intermediate, and high) according to the number of plasticity alleles carried by the adolescents was created. Using a multigroup approach, we examined whether the associations between physical activity and depressive symptoms differed between the three cumulative plasticity groups, as well as between the individual polymorphisms. Results: We found significant cross-sectional and cross-lagged paths from physical activity to depressive symptoms and vice versa. Neither the cumulative plasticity index nor the individual polymorphisms modified the strengths of these associations. Conclusion: Associations between adolescents’ physical activity and depressive symptoms are not modified by plasticity genes.