Reciprocal associations between physical activity, physical self-concept, somatic symptoms, and depression from adolescence to young adulthood: Disaggregating within- and between-person effects

Abstract

The mechanisms by which physical activity may influence depressive symptoms, and vice versa, during adolescence are not well understood. The present study examined the longitudinal, reciprocal within-person associations between physical activity and depressive symptoms among a community-based sample of adolescents who were followed into young adulthood, while also estimating between-person effects and exploring physical self-concept and somatic symptoms as potential mediators. Data were from the Victoria Healthy Youth Survey (V-HYS), which followed adolescents (W1; N = 662; ages 12–18) biannually for ten years into young adulthood (W6; n = 478; ages 22–29). Random-intercept cross-lagged panel models were specified to understand the within- and between-person associations between physical activity and depressive symptoms, as well as potential mediation by physical self-concept and somatic symptoms. Results showed anticipated between-person associations between physical activity, physical self-concept, somatic symptoms, and depressive symptoms. Although within-person increases in depressive symptoms predicted decreases in physical activity two years later, within-person deviations in physical activity did not significantly predict subsequent changes in depressive symptoms. Within-person cross-lagged associations between physical self-concept and depressive symptoms, as well as somatic symptoms and depressive symptoms, were significant and bidirectional in nature. Results are consistent with past research demonstrating the potential long-term and enduring health risks of depressive symptoms. Future research that uses a shorter timeframe between assessments (e.g., days or weeks) may further clarify the link between physical activity and depression, including potential mechanisms that explain why this association unfolds.