Effects of greenspace interventions on mental disorders - a systematic review and meta-analysis

This WHO-funded systematic review with meta-analysis summarizes the evidence of greenspace interventions on mental disorders, with patient-reported outcomes. The databases were searched through 23 October 2023: MEDLINE, Embase, PsycINFO and Cochrane Central Register of Controlled Trials. Randomized-controlled trials (RCT) for mental disorders with provider directed nature-based therapeutic interventions, or individual/group programs to improve health outcomes in nature were included. A search update was executed on Embase through 19 November 2025. The review was pre-registered in PROSPERO (CRD42023452769) and follows the PRISMA guidelines. Investigated interventions were detailed using the TIDieR Checklist. The study data were extracted, reviewed and scrutinized for risk of bias assessment by two reviewers independently. The meta-analysis was conducted with random-effects for all patient relevant outcomes. The mean differences (MD) or the standardized mean differences (SMD, Hedge’s g ) and their 95 % confidence intervals (CIs) were calculated between groups. The proportion of variability in effect estimates between studies that is not due to sampling error was categorized using I 2 , the between-studies variance in true effects was categorized using τ2. Further, we report prediction intervals. While twenty RCTs (1299 patients) were included, only twelve studies were eligible for quantitative data syntheses. All but one had a high or unclear risk of bias. Between two and maximum six studies could be included per meta-analysis. For all psychiatric indications, meta-analysis revealed a significant effect on depressive symptoms (SMD -0.52, p = .006, 95%CI [-0.89;-0.15]), anxiety (SMD -0.65, p = .005, 95%CI [-1.10;-0.20]) and positive affect (MD -1.70, p < .001, 95%CI [-2.66;-0.74]). Declines in positive affects may indicate a normalization of dysregulated or psychosis-related activations rather than a loss of positive emotions. Effects on negative affect, well-being and stress were not significant (all p ≥ 0.05). Among patients diagnosed with depression, significant improvement on depressive symptoms was evident (SMD -0.39, p = .049, 95%CI [-0.77;-0.001]). Among patients diagnosed with schizophrenia, significant improvements on positive affect (MD -1.75, p < .001, 95%CI [-2.73;-0.77]), negative affect (MD -1.39, p = .01, 95%CI [-2.38; -0.40]), and general psychopathology (MD -1.21, p < .001, 95%CI [-1.90;-0.53]) were evident. Heterogeneity differed and prediction intervals crossed zero. Overall, the current evidence is insufficient to establish efficacy. Future trials require larger samples, standardized interventions, longer follow-up, and improved methodological rigor to enable robust conclusions. • Greenspace interventions significantly reduce depressive symptoms, anxiety and positive affect • In schizophrenia, greenspace reduces positive and negative affect • Significant improvement seen with greenspace interventions in general psychopathology for schizophrenia • Greenspace interventions expand treatment options