The Effect of Exercise on Depressive Symptoms in Adolescents: A Systematic Review and Meta-Analysis

Tim Carter, Ioannis D. Morres, Oonagh Meade, Patrick Callaghan

The purpose of this review was to examine the treatment effect of physical exercise on depressive symptoms for adolescents aged 13 to 17 years. A systematic search of 7 electronic databases identified relevant randomized controlled trials. Following removal of duplicates, 543 texts were screened for eligibility. Screening, data extraction, and trial methodological quality assessment (using the Delphi list) were undertaken by 2 independent researchers. Standardized mean differences were used for pooling postintervention depressive symptom scores. Eleven trials met the inclusion criteria, 8 of which provided the necessary data for calculation of standardized effect size. Exercise showed a statistically significant moderate overall effect on depressive symptom reduction (standardized mean difference [SMD] = -0.48, 95% CI = -0.87, -0.10, p = .01, I(2) = 67%). Among trials with higher methodological scoring, a nonsignificant moderate effect was recorded (SMD = -0.41, 95% CI = -0.86, 0.05, p = .08). In trials with exclusively clinical samples, exercise showed a statistically significant moderate effect on depressive symptoms with lower levels of heterogeneity (SMD = -0.43, 95% CI = -0.84, -0.02, p = .04, I(2) = 44%). Physical exercise appears to improve depressive symptoms in adolescents, especially in clinical samples in which the moderate antidepressant effect, higher methodological quality, and lowered statistical heterogeneity suggest that exercise may be a useful treatment strategy for depression. Larger trials with clinical samples that adequately minimize the risk of bias are required for firmer conclusions on the effectiveness of exercise as an antidepressant treatment.