Effects of Forest Therapy on Depressive Symptoms: A Systematic Review and Meta-Analysis
Source: PubMed Central Open Access, NCBI / U.S. National Library of Medicine
Objective To systematically evaluate the efficacy of various forest therapy interventions for alleviating depressive symptoms using network meta-analysis, identify the optimal intervention pattern, and provide evidence-based guidance for clinical application. Eligibility Criteria Randomized controlled trials (RCTs) evaluating forest therapy for depressive symptoms were included. Studies were limited to English publications with extractable outcome data; non-forest interventions, simulated nature exposure, and studies with severe comorbidities or incomplete data were excluded. Information Sources PubMed, Medline, Embase, Cochrane Library, and CINAHL were searched from inception to May 2026. Study Appraisal and Synthesis Methods Study quality was assessed using the Cochrane ROB 2.0 tool. Network meta-analysis was performed with a random-effects model; heterogeneity was quantified using. Subgroup analyses were conducted by population, intervention mode, and assessment scale. Publication bias was evaluated using funnel plots and the trim-and-fill method. Results Eighteen RCTs with 2169 participants were included. Random-effects meta-analysis showed forest therapy significantly reduced depressive symptoms: SMD = −1.009, 95% CI (−1.411, −0.608),< 0.0001. Severe inter-study heterogeneity existed (= 90.1%, 95% CI [−2.765, 0.746],< 0.0001). Leave-one-out sensitivity analysis confirmed stable pooled SMD ranging from −1.083 to −0.880 after removing any single trial. Trim-and-fill test d
Abstract
Objective To systematically evaluate the efficacy of various forest therapy interventions for alleviating depressive symptoms using network meta-analysis, identify the optimal intervention pattern, and provide evidence-based guidance for clinical application. Eligibility Criteria Randomized controlled trials (RCTs) evaluating forest therapy for depressive symptoms were included. Studies were limited to English publications with extractable outcome data; non-forest interventions, simulated nature exposure, and studies with severe comorbidities or incomplete data were excluded. Information Sources PubMed, Medline, Embase, Cochrane Library, and CINAHL were searched from inception to May 2026. Study Appraisal and Synthesis Methods Study quality was assessed using the Cochrane ROB 2.0 tool. Network meta-analysis was performed with a random-effects model; heterogeneity was quantified using. Subgroup analyses were conducted by population, intervention mode, and assessment scale. Publication bias was evaluated using funnel plots and the trim-and-fill method. Results Eighteen RCTs with 2169 participants were included. Random-effects meta-analysis showed forest therapy significantly reduced depressive symptoms: SMD = −1.009, 95% CI (−1.411, −0.608),< 0.0001. Severe inter-study heterogeneity existed (= 90.1%, 95% CI [−2.765, 0.746],< 0.0001). Leave-one-out sensitivity analysis confirmed stable pooled SMD ranging from −1.083 to −0.880 after removing any single trial. Trim-and-fill test detected publication bias; after imputing seven missing studies, the therapeutic effect remained significant yet slightly weakened. Subgroup analyses found significant antidepressant effects in healthy people (SMD = −0.642, 95% CI [−0.981, −0.303],= 0.0002), chronic somatic disease patients (SMD = −1.431, 95% CI [−2.461, −0.400],= 0.0065) and depressed patients (SMD = −1.187, 95% CI [−1.662, −0.712],= 0.05), with no intergroup efficacy difference across populations (= 0.1961). Both guided (SMD = −0.881, 95% CI [−1.344, −0.418],= 0.0002) and unguided forest therapy (SMD = −1.134, 95% CI [−1.786, −0.482],= 0.0007) worked comparably. Conclusion Forest therapy is an effective non-pharmacological intervention for depressive symptoms. While forest therapy seems to offer benefits, the certainty of the evidence is constrained by significant heterogeneity, potential publication bias, a scarcity of direct comparisons among intervention models, and a lack of long-term follow-up studies. Future research should standardize protocols, conduct large-scale RCTs, incorporate long-term follow-up, and develop regionally adapted interventions. Registration PROSPERO CRD420261402570.