Effects of exercise on depression, anxiety, and physiological/biochemical markers in college students with depressive symptoms: A three-level meta-analysis

BACKGROUND: Depressive symptoms are increasingly common among college students, and exercise may be a promising intervention. This three-level meta-analysis examined the effects of exercise on depressive symptoms, anxiety symptoms, and physiological/biochemical markers in college students with depressive symptoms. METHODS: Five databases were searched from inception to October 2025 for randomized controlled trials of exercise interventions in this population. Risk of bias was assessed using RoB 2, and certainty of evidence was evaluated with GRADEpro. Effect sizes, moderator analyses, influence diagnostics, publication bias tests, and meta-regressions were conducted in R. RESULTS: Twenty-two studies involving 1255 participants were included. Low-certainty evidence suggested that exercise may reduce depressive symptoms (Hedges' g = -1.14, 95% CI [-1.45, -0.84], p < 0.001; 95% PI [-2.33, 0.05]). Low-certainty evidence also suggested a possible benefit for anxiety symptoms (g = -0.77, 95% CI [-1.28, -0.26], p = 0.008; 95% PI [-1.55, 0.01]), whereas very-low-certainty evidence suggested a possible benefit for physiological/biochemical markers (g = -0.97, 95% CI [-1.23, -0.71], p < 0.001). Sex composition and control group type significantly moderated depressive outcomes. Dose-response analyses suggested a possible nonlinear association between intervention duration and depressive symptom reduction, whereas total exercise time was not significant. CONCLUSIONS: Exercise may reduce depressive symptoms and may have potential benefits for anxiety symptoms and physiological/biochemical markers in college students with depressive symptoms. However, given methodological limitations and low-to-very-low certainty of evidence, these findings should be interpreted cautiously. Future well-designed RCTs are warranted.