Effectiveness of exercise on perinatal depression and anxiety symptoms: A network meta-analysis and dose-response analysis

Background: Perinatal depression and anxiety adversely affect maternal well-being. Although exercise is a promising strategy to improve mental health, the optimal prescription remains uncertain.

Objectives: This study compares the effectiveness of various exercise forms in alleviating depression and anxiety symptoms in perinatal women and examines the influence of exercise volume on intervention effectiveness.

Method: A comprehensive search strategy was conducted across PubMed, Cochrane Library, Embase, and Web of Science from inception to September 18, 2025. The search strategy focused on MeSH terms and keywords such as "Exercise," "perinatal," "depression," and "anxiety".

Selection criteria: Randomized controlled trials (RCTs) comparing the effectiveness of various types of exercise on perinatal depression and anxiety symptoms were included.

Data collection and analysis: Bayesian random-effects network meta-analysis was performed to evaluate the effectiveness of the interventions. Subgroup analyses were stratified by depression symptom presence in women and perinatal stage (prenatal versus postnatal). A Bayesian dose-response network meta-analysis modeled the nonlinear association between exercise volume and improvement in mental health outcomes.

Results: Of 8935 screened studies, 43 RCTs encompassing 3843 participants were included. Mind-body exercise (standardized mean difference [SMD] = -0.90, 95% credible interval [95% CrI]: -1.30, -0.59) and aerobic exercise (SMD = -0.60, 95% CrI: -1.10, -0.15) significantly improved perinatal depression symptoms. Specifically, Pilates (SMD = -1.50, 95% CrI: -2.70, -0.34) and yoga (SMD = -1.01, 95% CrI: -1.60, -0.46) demonstrated the most pronounced effects. Among women with depressive symptoms, aerobic exercise yielded the greatest benefit (SMD = -0.97, 95% CrI: -1.52, -0.46), followed by mind-body exercise (SMD = -0.44, 95% CrI: -0.82, -0.08). Stage-specific analyses revealed that aerobic (SMD = -0.86, 95% CrI: -1.40, -0.38) and mind-body (SMD = -0.58, 95% CrI: -0.89, -0.31) exercise were efficacious during the prenatal phase, whereas mind-body exercise exerted superior postnatal efficacy (SMD = -1.60, 95% CrI: -2.60, -0.70). The dose-response analysis indicated a nonlinear relationship, with a clinically important difference (SMD = -0.20) observed at approximately 100 MET-min/week. For perinatal anxiety symptoms, mind-body exercise also exhibited beneficial effects (SMD = -1.40, 95% CrI: -2.60, -0.19).

Conclusion: Both mind-body and aerobic exercises demonstrated substantial antidepressant effects, with Pilates and yoga eliciting the greatest benefits. Aerobic exercise showed the greatest efficacy for women with depressive symptoms. Aerobic and mind-body exercise improved prenatal depression symptoms, whereas mind-body exercise was more efficacious postpartum. The findings support stage-specific and volume-adjusted exercise prescriptions as viable nonpharmacological strategies for optimizing perinatal mental health outcomes.

anxiety; depression; dose–response; exercise; network meta‐analysis; perinatal.