The effects of physical exercise on anxiety symptoms of college students: A meta-analysis

The study addresses growing anxiety among college students, characterized by tension, worry, and uneasiness that may escalate to anxiety disorders. College students often face academic, interpersonal, and life stressors, with the COVID-19 pandemic exacerbating mental health issues and physical inactivity. Conventional treatments (medication and CBT) are effective but have side effects and relapse risks. Exercise has been proposed as a complementary, accessible intervention with physiological and psychological benefits, including mind–body practices (yoga, tai chi). The research aims to synthesize RCT evidence on whether physical exercise reduces anxiety symptoms in college students and to compare effects across exercise modes, intensities, durations, and frequencies to identify optimal parameters.

Prior observational and experimental research suggests an inverse association between physical activity—especially aerobic exercise—and anxiety symptoms. Aerobic exercise has been linked to improved cognitive performance, subjective wellbeing, and mental health. Mind–body practices like yoga and tai chi have also shown anxiolytic effects. Some studies indicate resistance training may have limited or mixed effects on state anxiety. Despite a broad literature, few studies directly compare aerobic exercise to other modalities for anxiety reduction, motivating this meta-analytic comparison across exercise types and dosing variables.

Design: Systematic review and meta-analysis of RCTs. Databases: CNKI, Wanfang, Weipu, PubMed, Embase, Cochrane Library, Web of Science; search up to January 19, 2022 using Boolean terms combining anxiety-related keywords, exercise/physical activity terms, and randomized trial design terms. Selection criteria (PICOS): Participants—college students; Interventions—aerobic exercise, yoga, tai chi, resistance training, or other physical activity; Comparators—control conditions; Outcomes—anxiety-related measures; Study design—randomized controlled trials. Exclusions: gray literature and reviews. Screening: PRISMA-guided; 3,808 records identified, 2,345 after deduplication, preliminary screening to 1,500, 97 full texts assessed, 9 RCTs included (n=483). Quality assessment: PEDro scale, independently scored by two authors with a third resolving disagreements; scores ranged 5–8 (average ~6), meeting criteria for random allocation, ITT analysis, between-group statistical analysis, and point/variance measurement; one study reported allocation concealment. Data synthesis: Continuous outcomes pooled using standardized mean difference (SMD) with 95% CI. Heterogeneity assessed via I²; random-effects model used if I² > 50%, otherwise fixed-effects. Stata 16.0 used for meta-analysis, subgroup analyses by exercise type, frequency, duration, and intensity, and publication bias assessed with Egger’s regression.

- Overall effect: Physical activity significantly reduced anxiety symptoms among college students (SMD = -0.55, 95% CI = -0.76 to -0.35; Z = 5.38; P < 0.001). Heterogeneity was moderate (I² = 63.5%; P < 0.05), leading to use of a random-effects model. - Publication bias: No evidence detected (Egger’s test t = 0.67; 95% CI = -4.69 to 8.34; P = 0.52). - Subgroup by exercise type: Aerobic exercise (SMD = -0.39; 95% CI = -0.74 to -0.04; P = 0.027; I² = 2.1%), yoga (SMD = -0.76; 95% CI = -1.14 to -0.39; P < 0.001; I² = 84.5%), and other exercise types (SMD = -0.50; 95% CI = -0.85 to -0.14; P = 0.006; I² = 90%) showed significant reductions; resistance exercise did not (SMD = -0.83; 95% CI = -1.74 to 0.09; P = 0.077). - Subgroup by frequency: High (≥5 times/week) (SMD = -1.14; 95% CI = -1.61 to -0.66; P < 0.001) and medium (3–4 times/week) (SMD = -0.57; 95% CI = -0.87 to -0.28; P < 0.001; I² = 71%) were effective; low frequency (1–2 times/week) was not (SMD = -0.22; 95% CI = -0.57 to 0.13; P = 0.217; I² = 0%). - Subgroup by duration: Medium (8–14 weeks) (SMD = -0.93; 95% CI = -1.32 to -0.54; P < 0.001; I² = 50%) and long (>14 weeks) (SMD = -0.57; 95% CI = -0.92 to -0.22; P = 0.001; I² = 91.8%) were effective; short (<8 weeks) was not (SMD = -0.29; 95% CI = -0.61 to 0.03; P = 0.080; I² = 0%). - Subgroup by intensity: High (SMD = -0.41; 95% CI = -0.76 to -0.05; P = 0.024; I² = 14.1%) and medium intensity (SMD = -0.84; 95% CI = -1.12 to -0.57; P < 0.001; I² = 59%) were effective; low intensity was not (SMD = 0.13; 95% CI = -0.40 to 0.66; P = 0.629).

The meta-analysis demonstrates that physical activity, particularly aerobic exercise and yoga, significantly alleviates anxiety symptoms among college students. These findings support exercise as a complementary strategy to established treatments (medication/CBT), offering accessible, low-risk options suitable for campus settings. The dose–response patterns observed suggest that medium to high intensity, moderate to high frequency, and interventions sustained over medium to longer durations yield more robust anxiolytic effects, informing practical program design. Resistance training did not show significant effects, aligning with prior mixed evidence on its anxiolytic potential. Overall, the results address the research question by identifying effective exercise modalities and dosing characteristics that can be implemented to reduce anxiety in this population, with relevance for university health services and mental health promotion initiatives.

Exercise interventions, notably aerobic exercise and yoga, effectively reduce anxiety symptoms in college students. Optimal results are associated with moderate to high intensity, higher frequency (≥3 times/week), and medium to longer durations (≥8 weeks). Incorporating structured aerobic programs into student routines could meaningfully alleviate anxiety and support overall physical and mental health. Future research should employ higher-quality, adequately powered RCTs with standardized anxiety measures to refine exercise prescriptions across modalities and intensities and to compare exercise as adjunctive versus standalone interventions.

- Population specificity: Findings are limited to university students and may not generalize to non-student young adults or specific subgroups (e.g., young women of childbearing age). - Limited number of RCTs: The small sample of included studies raises potential selection bias and limits precision. - Heterogeneity: Diverse outcome measures, exercise modalities, and dosing parameters contributed to moderate heterogeneity, complicating direct comparisons. - Resistance exercise evidence: Few studies and inconsistent results limit conclusions for resistance training. - Study quality: While methodological quality was adequate (mean PEDro ~6), allocation concealment and blinding were limited in several trials.