F1000Research F1000Research 2046-1402 F1000 Research Limited London, UK 10.12688/f1000research.166852.2 Systematic Review Articles Greenspace exposure and associated health outcomes: an updated systematic review of reviews

[version 2; peer review: 5 approved with reservations]

 Bryer Brittnee Data Curation Formal Analysis Investigation Methodology Validation Visualization Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0009-0009-2910-4843 1 Osborne Nicholas J Conceptualization Methodology Supervision Validation Writing – Review & Editing https://orcid.org/0000-0002-6700-2284 1 2 3 Wang Jialu Data Curation Investigation Validation Writing – Review & Editing 4 Dasgupta Rajarshi Validation Writing – Review & Editing https://orcid.org/0000-0003-0051-5090 5 Williams Gail Conceptualization Methodology Supervision Writing – Review & Editing 1 Darssan Darsy Conceptualization Data Curation Formal Analysis Investigation Methodology Project Administration Supervision Validation Visualization Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0001-7050-9150 a 1 The University of Queensland School of Public Health, Herston, Queensland, Australia School of Population Health, University of New South Wales, Sydney, New South Wales, Australia European Centre for Environment and Human Health, University of Exeter Medical School, Exeter, England, UK Queensland University of Technology School of Public Health and Social Work, Kelvin Grove, Queensland, Australia School of Public Policy, Indian Institute of Technology, Delhi, New Delhi, India d.darssan@uq.edu.au

No competing interests were disclosed.

 18 4 2026 2025 14 726 9 4 2026 Copyright: © 2026 Bryer B et al. 2026 This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background

Studies, including umbrella reviews, have provided evidence linking greenspace exposure to specific health outcomes. To address methodological limitations in existing umbrella reviews, this umbrella review aims to expand knowledge on the association between greenspace and human health by summarising evidence from both quantitative and qualitative systematic reviews and meta-analyses.

 Methods

This umbrella review builds upon our previous review with updates to reflect recent developments in the field. Five databases (PubMed, Embase, the Cumulative Index to Nursing and Allied Health Literature [CINAHL], Scopus, and the Cochrane Database of Systematic Reviews) were searched for articles published in English in peer-reviewed journals between December 2020 and June 2024. Systematic reviews and meta-analyses were included if they clearly defined measures of greenspace exposure and reported health outcomes directly attributable to greenspace exposure (PROSPERO: CRD42022383421). The methodological quality and risk of bias of each included review were evaluated by two independent reviewers.

 Results

A total of 45 articles were included in this umbrella review. Greenspace exposure was beneficially associated with mental health, general health and quality of life, and cardiovascular and metabolic health. Inconsistent associations were observed for respiratory health and allergies, maternal health and birth outcomes, and cancer. Most of the included reviews had a high risk of bias and critically low methodological quality.

 Discussion

This umbrella review highlights the multifaceted health benefits of greenspace exposure while underlining existing evidence gaps and methodological challenges. The results indicate that increasing access to greenspace could be a simple and cost-effective strategy to improve population health and reduce health inequalities.

 Human health natural environment outdoor environment umbrella review biodiversity greenery urbanisation The author(s) declared that no grants were involved in supporting this work. Amendments from Version 1

The revised manuscript incorporates the recommendations provided by the reviewers. Specifically, the descriptive detail in Tables 1 and 5 has been expanded, the corrected coverage area (CCA) has been calculated and reported, and the discussion of the study’s limitations has been extended. Summary sentences have also been added at the end of each health outcome subsection in the results to support clearer interpretation.

 Introduction

The rapid pace of urbanisation and environmental change has escalated interest in understanding how natural environments impact human health ( Farkas et al., 2023; Wang et al., 2024). Urban greenspaces, encompassing parks, forests, and other vegetated areas, have emerged as crucial components of urban ecosystems, offering potential health benefits ( Yang et al., 2021). Studies, including umbrella reviews, have provided evidence linking greenspace exposure to specific health outcomes, such as mental ( Bonaccorsi et al., 2023; Cuijpers et al., 2023), cognitive ( Zare Sakhvidi et al., 2023), and cardiometabolic health ( Liu et al., 2023; Sharifi et al., 2024), as well as birth outcomes ( Khalaf et al., 2025; Zare Sakhvidi et al., 2023), mortality ( Bryer et al., 2024; Song et al., 2024), and quality of life (QoL) ( Bonaccorsi et al., 2023). However, limitations in the methodology of these reviews have resulted in some gaps in the literature.

Many systematic reviews struggle to quantitatively synthesise evidence on the association between greenspace and individual health outcomes due to the limited number of original studies available. In such cases, these reviews often present their findings narratively. Conversely, in 2024, Xie et al. (2024) conducted an umbrella review examining the credibility of evidence on the association between greenspace and human health. While the authors graded the credibility of the current evidence, they focused solely on quantitative results. When conducting a systematic review on the association between greenspace and human health, authors should consider both quantitative and qualitative results. This will provide a more comprehensive understanding of the association and enhance the generalisability of the findings.

Recent years have also seen the emergence of new umbrella reviews on the greenspace-health association. However, these reviews have typically focused on a narrower selection of health domains or demographic groups. For instance, some reviews explored the association between greenspace and specific health outcomes, such as cardiovascular disease (CVD) ( Liu et al., 2023) and mental health ( Cuijpers et al., 2023). Others have concentrated on specific age groups, including children and adolescents ( Zare Sakhvidi et al., 2023) and older adults ( Bonaccorsi et al., 2023).

There is a clear need for an updated and comprehensive review that outlines both quantitative and qualitative evidence while encompassing all health outcomes across all age groups. To address this gap, this umbrella review aims to expand existing knowledge on the association between greenspace and human health by summarising evidence from both quantitative and qualitative systematic reviews and meta-analyses.

 Methods

This umbrella review builds upon our previous review (January 2010 – December 2020) ( Bryer et al., 2024) with updates to reflect recent developments in the field. The protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO; registration ID: CRD42022383421) ( Bryer et al., 2022). The review adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines ( Moher et al., 2009).

Search strategy

The existing search strategy from our previous review was used to identify peer-reviewed systematic reviews and meta-analyses examining the association between greenspace exposure and human health. The search was restricted to studies conducted in humans published between December 2020 and June 2024, including reviews first published online during this period (the full search strategy for each database is available in the extended data: Supplementary Table S1). Searches were conducted on July 1, 2024 across five databases: PubMed, Embase, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), Scopus, and the Cochrane Database of Systematic Reviews. EndNote was used for reference management ( The EndNote Team, 2013).

 Eligibility criteria

Two authors independently screened the titles and abstracts and full texts of identified systematic reviews. Inclusion criteria were as follows: i) systematic review or meta-analysis; ii) published in English in a peer-reviewed journal; iii) greenspace exposure clearly defined using objective or subjective measures; iv) reported health outcome(s) directly attributable to greenspace exposure (e.g., independent of blue space). Reviews were excluded if they were: i) systematic reviews that did not adhere to standardised systematic review methodologies (e.g. lacking database search details, search terms, or quality assessment); ii) scoping reviews; iii) reviews that combined greenspace effects with blue space; iv) studies that only reported on determinants of health (e.g. body mass index [BMI], physical activity, social health). Disagreements during screening were resolved through discussions with the last author. Covidence software was used to conduct all stages of screening ( Covidence systematic review software, 2022).

 Data extraction

Two authors independently extracted data using a pre-established data extraction form ( Bryer et al., 2024). Extracted data included: design of original studies, types and measures of greenspace exposures, health outcomes, and the main findings of the review. Any discrepancies in extracted data were resolved through discussions with the last author.

 Analysis criteria

The objectives, greenspace measures, health outcomes, and findings of each included review were summarised qualitatively. Consistent with the purpose of umbrella reviews, which outline evidence from systematic reviews rather than re-synthesising the results, no meta-analysis was performed ( Aromataris et al., 2015; Pollock et al., 2024). Where quantitative data were available, effect estimates and 95% confidence intervals were reported.

The total number of original studies, classified by study design, was recorded, along with details of greenspace measures and reported health outcomes. For each health outcome, the corresponding International Classification of Diseases, 10 th Edition (ICD-10) code(s) was also noted ( World Health Organization, 2019).

Methodological quality and risk of bias were independently evaluated by two authors using A MeaSurement Tool to Assess systematic Reviews 2 (AMSTAR-2) ( Shea et al., 2017) and the Risk of Bias Assessment Tool for Systematic Reviews (ROBIS) ( Shea et al., 2017), respectively. Disagreements were resolved through discussions with the last author. Systematic reviews were not excluded based on methodological quality or risk of bias.

 Results Study selection

The search strategy identified 1,981 unique articles, of which 277 were retained for full-text screening. Ultimately, 45 articles met the inclusion criteria for this umbrella review. The PRISMA flowchart is available under the Reporting guidelines section at the end of this article.

Study type and characteristics

Most included systematic reviews were published in 2022 (n = 18, 40.0%). The number of original studies within these reviews ranged from two to 140, with participant sample sizes spanning from two to 159 million ( Table 1). Various study designs were represented, with cross-sectional studies being the most prevalent (n = 501, 45.7%) ( Table 2).

Characteristics of included systematic reviews <xref ref-type="table-fn" rid="tfn1">*</xref>.
Author (year) No. original studies (location) Type of study (n) Objective(s) of the study Sample size (age range) Greenspace measures Outcome measures Main finding(s)
Aarthi et al. (2023) 3 (Asia) Cross-sectional (3) Review associations of built environment characteristics with T2DM in Asia. 4,155-341,211 (0-99) NDVI, SAVI, greenspace ratio, green vision index, evergreen tree configuration T2DM, glucose levels Greenness associated with lower glucose levels in Taiwan and China.
Ahmer et al. (2024) 31 (North America, Europe, Asia, Oceania) Cross-sectional (26), cohort (3), case-control (1), quasi-experimental (1) Review influence of residential greenspaces on birth outcomes of pregnant women. 301-3,753,788 (unclear) NDVI, greenspace/land percentage, tree canopy coverage, LiDAR BW, LBW, PTB, SGA Residential greenspaces positively associated with increased BW and lower odds of LBW, PTB, and SGA deliveries.
Batterham et al. (2022) 4 (North America, Europe, Asia, Oceania) Cross-sectional (4) Synthesise quantitative evidence for relationship between environmental factors and mental health in rural areas. 2,020-33,823 (unclear-89) Percentage of greenspace type, land cover classes, proportion of forest & agricultural areas, total greenspace ratio General mental health, anxiety, depression, affective disorders, schizophrenia Areas with greater greenspace might have better well-being, marginally less depression, and reduced hospitalisations.
Bianconi et al. (2023) 36 (North America, South America, Europe, Asia, Oceania) Ecological (15), cohort (13), cross-sectional (7), case-control (1) Assess impacts of urban greenery on cardiovascular and cerebrovascular disease morbidity and mortality. 469-43,000,000 (unclear-89) NDVI, SAVI, LAI, greenspace percentage/urban park percentage, distance from city parks larger than 1 hectare, tree canopy percentage CVD mortality, IHD mortality, stroke mortality, CHD, stroke, CVD, AMI, HF, heart disease, IHD Suggests greenspace exposure has a beneficial effect on cardiovascular and cerebrovascular health in urban settings.
Bolanis et al. (2024) 23 (North America, Europe, Asia, Africa, Oceania) Ecological (14), cross-sectional (4), longitudinal (3), pre-post design (1), randomised cross-over (1) Review evidence on association between greenspace exposure and suicide-related outcomes. 15-8,741,021 (11-77) NDVI, land cover map, greenspace percentage, number of parks per 1000 people, per capita park area, greenspace fragmentation, mean green patch area, green patch distance, percentage tree cover, tree canopy coverage, activities in forest settings, mountain hiking, decrease time in greenspace Suicide mortality, self-harm, suicidal ideation Greenspace exposure may have a protective role across the entire spectrum of suicide outcomes with larger putative benefits observed among females.
Briggs et al. (2022) 20 (North America, Europe, Asia) RCT (20) Evaluate effectiveness of group-based gardening interventions for increasing well-being and reducing mental ill-health in adults. 20-89 (18-78) Gardening interventions Depression, anxiety, stress, QoL, overall mental health, mood, affect, PTSD, burnout Mixed evidence for the effectiveness of group-based gardening interventions for mental health and wellbeing.
Browning et al. (2022) 37 (North America, Europe) Cross-sectional (29), longitudinal (8) Consolidate evidence on how urbanicity modifies relationship between greenspace and health. 1,542-97,574,613 (0-85) NDVI, green landcover, public greenspace, distance to greenspace, distance-based geographic area, container-based geographic area Respiratory health, cardiovascular health, diabetes, mortality, cancer, birth outcomes More analyses showed stronger protective associations in urban areas than in less urban areas.
Cao et al. (2023) 14 (North America, Europe, Asia, Africa, Oceania) Cohort (10), cross-sectional (3), ecological (1) Determine association between greenness and AR in children and adolescents. 522-642,313 (0-17) NDVI, EVI, greenspace coverage, MODIS, green score (residential greenness and neighbourhood green land use) AR Results indicate no association between greenness exposure and AR in children and adolescents.
Ccami-Bernal et al. (2023) 13 (North America, Europe, Asia, Oceania) Cohort (13) Summarise evidence on exposure to greenspace and incidence of T2DM from longitudinal studies. 1,700-1,922,545 (15-87) NDVI, forest percentage, percentage of any greenspace, percentage of neighbourhood greenness, EVI, proportion of greenspace within1000m of the residence, VCF T2DM Exposure to greenspace could be a protective factor for the development of T2DM.
Chae et al. (2021) 13 (Asia) Pre-post (10), RCT (3) Analyse effects of forest therapy on immune function. 11-61 (unclear-80) Forest therapy Immune function Forest therapy programs may contribute to the improvement of immune function.
Chen et al. (2022) 2 (North America, Europe, Asia, Oceania) Qualitative (1), mixed methods (1) Identify types, benefits, motivations, and drawbacks of environmental volunteerism. 5-19 (unclear-92) Greening volunteerism Depression, positive affect, distress Environmental volunteerism can be considered a good model for productive aging and nature conservation.
De la Fuente et al. (2021) 19 (North America, Europe, Asia, Oceania) Cross-sectional (12), cohort (6), longitudinal (1) Know the magnitude of effect of greenspace on prevalence of T2DM through analysis of recent evidence. 76-3,920,000 (15-85) Greenspace exposure by postcode, NDVI, area of parks in neighbourhood, density of trees & park area, frequency & intensity of exposure to nature, street tree density, greenspace satisfaction, distance to nearest park T2DM There is evidence for the protective role of greenspaces in the urban context against T2DM.
Gao et al. (2024) 16 (North America, Europe, Asia) Cross-sectional (6), cohort (5), ecological (2), case-control (1), RCT (1) non-RCT (1) Investigate impact of greenspace on COPD-related disease outcomes and summarise reasons for heterogeneity. 18-2,185,170 (unclear) NDVI, green land cover, tree canopy cover, forest cover, MSAVI, forest bathing COPD Greenery may positively impact COPD-related health outcomes, although some studies have yielded different results.
Hsueh et al. (2022) 3 (North America, Europe, Oceania) Mixed methods (2), qualitative (1) Synthesise literature on community place-based interventions addressing loneliness and mental health problems. 13-56 (9-98) Community gardening, exposure to greenery Depression, anxiety, PTSD, overall mental health Insufficient evidence of an association between community-based interventions and loneliness or mental health problems.
Kang et al. (2022) 6 (Europe, Asia) RCT (6) Overview of RCTs on effects of forest-based interventions on mental health. 33-84 (unclear-79) Forest therapy, forest bathing, forest videos, forest walking Depression, anxiety, positive emotions, negative emotions, overall mental health Forest-based interventions improved mental health of participants in the intervention groups compared to those in the control groups.
Lampert et al. (2021) 8 (North America, Europe, Asia) Cross-sectional (8) Synthesise the literature about physical and mental health outcomes associated with community gardening. 50-332 (25-88) Community gardening Overall health, number of chronic conditions, stress, anxiety, depression, QoL, PTSD, overall mental wellbeing Community gardens are associated with health gains for their users, irrespective of age, being an affordable and efficient way of promoting physical and mental health and well-being.
Li et al. (2023) 14 (North America, Europe, Asia) Cohort (14) Integrate evidence from cohort studies to obtain best available epidemiological evidence on greenspace exposure and cancer. 7,300-10,481,566 (unclear) NDVI, proximity to greenspace All-site cancer, prostate cancer, skin cancer, colorectal cancer, breast cancer, lung cancer, bladder cancer, mouth cancer, other cancer, all-site cancer mortality, lung cancer mortality, prostate cancer mortality Greenspace exposure reduces lung cancer and prostate cancer mortality, and there may be a potentially beneficial association for prostate, lung, and breast cancer incidence.
Li and Lange (2023) 15 (North America, Europe, Asia, Oceania) Cross-over (6), RCT (3), cross-sectional (2), pre-post design (2), real-time (1), cohort (1) Provide a comprehensive analysis of impact of urban landscapes and interventions on stress reduction. 11-164 (14-85) Percentage land use, percentage of trees, residential gardens, natural/green corridor, semi-natural/natural areas, private greenspace, recreation civic space, parks, window green view, street greenery, landscape elements Psychological stress, mood, mental wellbeing, cognition, physiological stress, overall health Suggests that urban landscapes can provide therapeutically relevant changes in psychological states, physiological activity, and cognitive functioning.
Masdor et al. (2023) 5 (North America, Europe) Cohort (5) Investigate association between exposure to greenspace in adult population and outcomes of colorectal cancer. 19,408-2,441,566 (30-69) NDVI, garden percentage, natural environment percentage, ecosystem count, recreational facilities & parks count, residential proximity to greenspace, surrounding greenness Colorectal cancer, colorectal cancer mortality There is currently no link between colorectal cancer and greenspace.
Patwary et al. (2024) 18 (North America, Europe, Asia, Oceania) Cross-sectional (8), cohort (7), ecological (3) Assess association of greenspaces and built environments with MetS risk. 75-49,893 (10-112) NDVI, EVI, SAVI, VCF, distance to greenspace, tree canopy density, parkland percentage, percentage land use for park & recreation MetS Suggests potential benefit of greenspace in reducing MetS risk.
Piva et al. (2024) 10 (Asia) Exploratory (6), RCT (2), non-RCT (2) Explore physiological and psychological benefits of forest therapies on healthy and elderly populations. 18-88 (60-80) Forest bathing, forest exercise HR, HRV, BP, lung function, cortisol level, mood, depression, QoL, stress, cognitive impairment, neurochemical response Forest therapy interventions are effective in improving physiological, neurochemical, and biological parameters, and psychological well-being.
Qiu et al. (2022) 29 (Europe, Asia) Crossover RCT (15), parallel RCT (10), non-RCT (4) Toassess the effects of forest therapy on BP and SCC in urban residents. 11-348 (19-80) Forest therapy, forest walking, forest viewing SBP, DBP, SCC Longer forest therapy programs have greater BP- and SCC-lowering effects than shorter programs.
Rahimi-Ardabili et al. (2021) 73 (Asia) Cross-sectional (42), cohort (9), RCT (9), ecological (7), case-control (3), before-after study (2), quasi-experimental (1) Synthesise the emerging evidence on greenspace and health in mainland China. 20-159,000,000 (16-80) NDVI, SAVI, proximity/frequency of visits to parks, viewing greenery, green walking, percentage green space coverage, time spent in parks, land cover map Mood, tension-anxiety, depression, fatigue, overall mental health/mental wellbeing, stress, attention, affective states, ADHD, QoL, overall health, diabetes, eczema, number of chronic conditions, BP, hypertension, CHD, cardiovascular health, MetS, CVD, stroke, cognition, asthma, lung cancer, COPD, rhinitis, number of respiratory diseases, all-cause mortality, other causes of mortality, malaria, dysentery, dengue, tuberculosis, BW, miscarriage Positive associations found between greenspace and mental health, cardiovascular outcomes, and general health.
Ricciardi et al. (2022) 25 (North America, Europe, Asia, Oceania) Cross-sectional (19), longitudinal (6) Summarise studies using objective measures of greenspace exposure and cognitive functioning. 108-38,327 (unclear) NDVI, EVI, VFC, MEDIx, distance to NOE, percentage of parkland, total land area, total soft surface, tree canopy cover, time spent in green spaces, grass/shrub cover, average percent canopy cover Cognition, memory, attention, executive function Evidence was inconsistent but suggestive of a beneficial role of greenspace exposure on cognitive functions.
Rigolon et al. (2021) 90 (North America, South America, Europe, Asia, Oceania) Cross-sectional (79), longitudinal (11) Review whether green space can contribute to limiting health disparities and moving toward health equity. 106-97,574,613 (unclear-111) Gardening, green land cover, nature-based programs, exposure to public green space Eczema, asthma, diabetes, prostate cancer, CVD, all-cause mortality, PTB, infant mortality, overall health Greenspace had greater protective effects on overall health for low-SES people and neighbourhoods than for more affluent groups.
Rojas-Rueda et al. (2021) 19 (South America) Cross-sectional (14), ecological (3), longitudinal (2) Review epidemiological evidence on green spaces and health outcomes in Latin America. 120-103,000,000 (0-98) NDVI, frequency/duration of green space exposure, density/proximity to green space, presence of green space, biodiversity index, sustainable development index, park coverage Depression, overall mental well-being, cognitive functioning, QoL, infant mortality, life expectancy at birth, cardiorespiratory hospital admission, disability, overall health Most evidence suggests a positive association between green spaces and health in the region.
Siah et al. (2023) 36 (North America, Europe, Asia) RCT (22), quasi-experimental (7), crossover design (7) Identify and synthesise evidence on forest bathing and its impacts on individual well-being. 12-585 (18-90) Forest bathing Depression, anxiety, QoL, cortisol level, psychosocial well-being, direct-attention, stress, SBP, DBP, HR, HRV, cholesterol, inflammation, blood glucose, CVD risk, pain, disability, working memory, post-stroke fatigue There is evidence for the beneficial effects of forest bathing on psychological well-being.
Sivak et al. (2021) 22 (North America, South America, Asia) Cross-sectional (7), quasi-experimental (5), RCT (4), experimental (2), qualitative (3), case-control (1) Consolidate literature about vacant lots’ effects on human health, summarise findings and identify gaps in existing evidence. 12-49,246 (8-82) Exposure to vacant lots Stress, depression, overall mental health, BP, heavy metal contamination, dermatophytosis, toxocara, foodborne pathogens, West Nile Virus, physical injury Depression, stress, physical activity, relaxation and socialisation in outdoor areas, and HR improved when living near or being within view of a greened vacant lot versus an unmaintained control lot.
Tang et al. (2023) 35 (North America, Europe, Asia) Cohort (18), cross-sectional (10), ecological (4), case-control (3) Examine correlation between greenness exposure and chronic respiratory health issues. 478-10,481,566 (0-89) NDVI Asthma, AR, COPD, lung cancer, COPD mortality, lung cancer mortality There is a correlation between higher greenness exposure and a reduced risk of asthma incidence, lung cancer incidence, and COPD mortality.
Tharrey and Darmon (2021) 15 (North America, Europe, Asia) Cross-sectional (13), post-test only (1), RCT (1) Review quantitative studies analysing relationship between participation in collective gardens and gardeners’ health status in urban free-living adults in high-income countries. 30-13,133 (18-94) Community gardening, allotment gardening Mood, psychological well-being, psychological distress, stress general mental health, BP, lung function, general health Several studies found a positive association between collective garden participation and physical, mental, or social health, but the results came mostly from cross-sectional studies.
Tu (2022) 19 (North America, Europe, Asia) RCT (19) Explore effect of horticultural therapy on mental health. 13-150 (10-95) Horticultural therapy Depression, anxiety, stress, PTSD, positive affect, negative affect, overall mental wellbeing, memory Meta-analysis provided evidence that horticultural therapy has a positive effect on mental health.
Wang et al. (2022) 48 (North America, South America, Europe, Asia, Africa, Oceania) Cross-sectional (18), cohort (16), ecological (7), case-control (4), longitudinal (3) Review of evidence to elucidate association between greenness exposure and allergic diseases. 126-10,500,000 (0-73) NDVI, VCG, EVI, SAVI, GVI, LiDAR, generalised land use database, ArcGIS, land cover classification maps, distance to nearest city part, residential proximity to greenspace, perceptions of greenspace quality, LUG, percentage greenspace, total number of natural land-cover types, plant diversity, tree canopy cover & agricultural cover Asthma, AR, ARC, AD, food allergies Exposure to greener environments early in life may be a protective factor for AR and asthma in childhood.
Wu et al. (2022) 21 (North America, Europe, Asia, Oceania) Cohort(10), cross-sectional (7), case-control (4) Review association between greenspace exposure and incidence of asthma and AR. 187-59,754 (0-34) NDVI, LiDAR, CLC, GVI Asthma, AR Insufficient evidence of an association between greenspace exposure and asthma or AR.
Ye et al. (2022) 140 (North America, South America, Europe, Asia, Oceania) Cross-sectional (83), cohort (40), ecological (8), longitudinal (4), case-control (4), combined cohort cross-sectional (1) Summarise measures of greenspace, compile evidence of an association between exposure to greenspace and early onset of health outcomes in childhood and adolescence and explore research gaps on how greenspace affects human health at this early stage of life. 61-5,262,265 (0-19) NDVI, EVI, SAVI, greenspace coverage (rate, proportion, & percentage), proximity to greenspace, perceived greenspace quantity and quality, frequency/duration of greenspace use, street view green index Depression, anxiety, ADHD, overall mental well-being, autism, stress, psychological distress, QoL, attention, cognitive & motor function, cognitive development, working memory, asthma, rhinitis, AR, lung function, bronchitis, pneumonia, aeroallergen sensitisation, atopic sensitisation, BP, cardiometabolic health, allosteric load, overall health, myopia, astigmatism, intestinal parasites, IBD, leukaemia Though inconsistencies remain, protective effects have been reported with greenspace contacts on childhood mental health. Limited but promising findings also indicate that greenspace could be beneficial for children’s lung function and circulatory health, and reducing the prevalence of myopia.
Yi et al. (2022) 17 (Europe, Asia) RCT (17) Analyse effectiveness of forest therapy for both physiological and psychological areas. 11-144 (unclear-80) Forest therapy, nature meditation, viewing greenery, green exercise Depression, anxiety, mood, overall mental health, QoL, BP, HR, physiological stress, overall health Forest therapy reduced depressive symptoms, but there was insufficient evidence of an association with BP.
Yuan et al. (2021) 22 (North America, Europe, Asia, Oceania) Cohort (17), cross-sectional (5) Synthesise evidence from observational studies to assess relationships of greenspace exposure with mortality and cardiovascular outcomes in older individuals. 1,084-5,988,606 (25-unclear) NDVI, availability of greenspace, number of parks in buffer area, loss of trees, proportion of greenspace, distance to nearest greenspace, greenspaces per inhabitant, frequency/duration of visits to greenspace, tree canopy CVD, AMI, stroke, heart attack, CHD, IHD, all-cause mortality, non-accidental mortality, cancer mortality, stroke mortality, respiratory disease mortality, circulatory disease mortality, IHD mortality, AMI mortality, CVD mortality, infectious/parasitic disease mortality, CVD mortality, neurodegenerative disease mortality Some evidence for associations between increased greenness exposure and reduced risk of all-cause and stroke mortality, as well as major CVD outcomes in elderly populations.
Zagnoli et al. (2022) 12 (North America, Europe, Asia, Oceania) Cross-sectional (4), cohort (4), case-control (1), ecological (1), combined cohort cross-sectional (1), combined cohort case-control (1) Assess the relationship between environmental greenness and the risk of dementia and cognitive impairment. 2,424-1,737,460 (unclear) NDVI, land cover, land use, total greenspace, tree canopy Dementia, Alzheimer's dementia, non-Alzheimer's dementia, ADRD, VaD, PD, MS, cognitive impairment, dementia mortality, neurodegenerative disease mortality Living in a place with an intermediate greenness value may protect against dementia.
Zare Sakhvidi, Knobel, et al. (2022a) 29 (North America, Europe, Asia, Oceania) Cross-sectional (17), longitudinal (11), case-control (1) Review association between long-term exposure to greenspace and behavioural problems in children and explore sources of heterogeneities. 169-814,689 (2-18) NDVI, SAVI, MSAVI, duration of greenspace use, percent coverage, distance to greenspace, type of greenspace, quality of greenspace ADHD, externalising disorders, internalising disorders, hyperactivity, attention Determinants: conduct problems, peer problems, prosocial behaviour, conditional problems There are beneficial associations between exposure to greenspace with several behavioural outcomes in children.
Zare Sakhvidi, Yang, et al. (2022b) 18 (North America, Europe, Asia, Oceania) Cross-sectional (8), cohort (6), case-control (4) Assess evidence on association between exposure to greenspace and cancer incidence, prevalence, and mortality in adults. 1,910-28,600,000 (0-unclear) NDVI, proximity to greenness, land cover Skin cancer, colorectal cancer, breast cancer, prostate cancer, lung cancer, non-melanoma cancer, mouth & throat cancer, all cancer, brain tumour, lung cancer mortality, prostate cancer mortality, oesophageal cancer mortality, all cancer mortality Part of the available literature is suggestive of a possible protective association of greenspace exposure with breast and prostate cancer, while for other cancers the evidence is still very limited and heterogeneous.
Zhang et al. (2021) 26 (North America, South America, Europe, Asia, Oceania) Cross-sectional (24), longitudinal (2) Review pathways linking objectively measured greenspace exposure and mental health outcomes. 109-1,930,048 (12-91) NDVI, greenspace exposure via GIS, national land cover classification database, distance to greenspace, street view greenness, SVG Depression, anxiety, psychological distress, overall mental wellbeing Supportive evidence was found for a direct pathway from greenspace exposure to mental illness and mental well-being.
Y. Zhang et al. (2024a) 60 (North America, South America, Europe, Asia, Oceania) Cross-sectional (32), cohort (28) Explore effects of greenspace exposure on common psychiatric disorders. 322-61,662,472 (2-97) NDVI, area of greenspace, parks, view of greenery, greenspace accessibility/availability Depression, anxiety, ADHD, schizophrenia, dementia, psychiatric disorders Greenspace linked to lower odds of depression, anxiety, dementia, schizophrenia, and ADHD.
Y. D. Zhang et al. (2024b) 20 (North America, South America, Europe, Asia, Oceania) Cross-sectional (10), non-RCT (5), RCT (2), before-after intervention (2), cohort (1) Comprehensively synthesise observational and interventional evidence on association between greenspace exposure and human microbiota. 2-9,129 (0-84) NDVI, EVI, percentage agricultural land & forest, percentage trees & grassland, percentage greenspace, green remediation, natural environment around home, rubbing hand with soil- & plant-based materials, direct exposure to soil & plants, gardening Gut microbiota, skin microbiota, oral microbiota, nasal microbiota Greenspace exposure may diversify gut and skin microbiota andalter their composition to healthier profiles.
X. Zhang et al. (2022c) 24 (North America, Europe, Asia) Randomised crossover (12), non-randomised crossover (7), randomised parallel group (3), 2x2 factorial design (1), single group crossover (1) Conduct a systematic review of the anxiety-alleviation benefits of exposure to the natural environment 11-498 (unclear-79) Trail surrounded by trees, roads surrounded by trees, forest exposure, green area at university, park, garden & landscape photographs Anxiety Natural environment or pictures and videos of the natural environment can be effective in reducing anxiety.
Zhao et al. (2022) 38 (North America, South America, Europe, Asia, Oceania) Cross-sectional (27), cohort (9), longitudinal (1), case-control (1) Review associations between greenspace and BP levels/hypertension. 73-3,920,000 (4-111) NDVI, SAVI, EVI, greenspace percentage in buffer, tree cover, distance to greenspace, land cover, number of street trees, proportion greenspace, availability of greenspace Hypertension, SBP, DBP Greenspace may play an important role in decreasing BP levels and the prevalence of hypertension.
Zhao et al. (2021) 9 (North America, Europe, Asia) Cohort (4), case-control (3), cross-sectional (2) Summarise natural, physical, and social environmental factors associated with cognitive impairment and dementia. 675-1,737,460 (34-101) NDVI, generalised land use, density of greenspace, percentage of greenspace, distance to greenspace Dementia, Alzheimer’s dementia, non-Alzheimer’s dementia, cognitive impairment More residential greenness might be favourable in the risk of cognitive impairment and dementia.

AD: atopic dermatitis; ADHD: attention deficit hyperactivity disorder; ADRD: Alzheimer’s disease and related dementia; AMI: acute myocardial infarction; AR: allergic rhinitis; ARC: allergic rhino-conjunctivitis; BMI: body mass index; BP: blood pressure; BW: birthweight; CHD: coronary heart disease; COPD: chronic obstructive pulmonary disease; CLC: coordination of information of the environment (CORINE) land cover; CVD: cardiovascular disease; DBP: diastolic blood pressure; EVI: enhanced vegetation index; GIS: geographic information system; GVI: green view index; HF: heart failure; HR: heart rate; HRV: heart rate variability; IBD: irritable bowel disease; IHD: ischaemic heart disease; LAI: leaf area index; LBW: low birth weight; LiDAR: light detection and ranging; LUG: land use greenness; MEDIx: multiple environmental deprivation index; MetS: metabolic syndrome; MODIS: moderate resolution imaging spectroradiometer; MS: multiple sclerosis; MSAVI: modified soil adjusted vegetation index; NDVI: normalised difference vegetation index; NOE: natural outdoor environment; PD: Parkinson’s disease; PTB: pre-term birth; PTSD: post-traumatic stress disorder; QoL: quality of life; RCT: randomised controlled trial; SAVI: soil-adjusted vegetation index; SBP: systolic blood pressure; SCC: serum cortisol concentration; SES: socioeconomic status; SGA: small for gestational age; SVG: street-view imagery-based greenness indices; T2DM: type 2 diabetes mellitus; VaD: vascular dementia; VCF: vegetative continuous field; VFC: vegetation fractional cover.

 Study design of original studies in included systematic reviews.
Study design n (%)
Cross-sectional 501(45.7%)
Cohort 211 (19.2%)
Randomised Controlled Trial (RCT) 109 (9.9%)
Ecological 65 (5.9%)
Longitudinal 52 (4.7%)
Case-control 32 (2.9%)
Other a 127 (11.6%)

Other: crossover RCT, quasi-experimental, pre-post design, crossover design, randomised crossover, non-RCT, parallel RCT, non-randomised crossover, exploratory, qualitative, before-after study, randomised parallel group, mixed methods, experimental, combined cohort cross-sectional, post-test only, single group crossover, real-time study, randomised controlled crossover, 2x2 factorial design, combined cohort case-control.

 Greenspace exposure

A wide array of greenspace exposure measures were reported ( Table 3). The Normalised Difference Vegetation Index (NDVI), a common quantitative measure of vegetation, appeared in 27 systematic reviews (60.0%). Other frequently used measures included greenspace percentage, proximity to greenspace, presence of tree canopies, Soil Adjusted Vegetation Index (SAVI), Enhanced Vegetation Index (EVI), availability of greenspace, and gardening. Additionally, 37 reviews (82.2%) reported alternative greenspace measures, encompassing both qualitative and quantitative assessments, as well as intervention activities.

Measures of greenspace exposure in included systematic reviews.
Greenspace exposure measure Counts of greenspace exposure measure (% out of 44 reviews) Review articles
NDVI b 27 (60.0%) Aarthi et al. (2023); Ahmer et al. (2024); Bianconi et al. (2023); Bolanis et al. (2024); Cao et al. (2023); Ccami-Bernal et al. (2023); De la Fuente et al. (2021); Gao et al. (2024); Li et al. (2023); Li and Lange (2023); Masdor et al. (2023); Patwary et al. (2024); Rahimi-Ardabili et al. (2021); Ricciardi et al. (2022); Rojas-Rueda et al. (2021); Tang et al. (2023); Wang et al. (2022); Wu et al. (2022); Ye et al. (2022); Yuan et al. (2021); Zagnoli et al. (2022); Zare Sakhvidi, Knobel, et al. (2022a); Zare Sakhvidi, Yang, et al. (2022b); Zhang et al. (2021); Y. D. Zhang et al. (2024b); Zhao et al. (2022); Zhao et al. (2021)
Greenspace percentage 16 (35.6%) Ahmer et al. (2024); Batterham et al. (2022); Bianconi et al. (2023); Bolanis et al. (2024); Ccami-Bernal et al. (2023); Li and Lange (2023); Masdor et al. (2023); Patwary et al. (2024); Rahimi-Ardabili et al. (2021); Ricciardi et al. (2022); Wang et al. (2022); Ye et al. (2022); Zare Sakhvidi, Knobel, et al. (2022a); Y. D. Zhang et al. (2024b); Zhao et al. (2022); Zhao et al. (2021)
Proximity to greenspace 16 (35.6%) Bianconi et al. (2023); De la Fuente et al. (2021); Li et al. (2023); Masdor et al. (2023); Patwary et al. (2024); Rahimi-Ardabili et al. (2021); Ricciardi et al. (2022); Rojas-Rueda et al. (2021); Wang et al. (2022); Ye et al. (2022); Yuan et al. (2021); Zare Sakhvidi, Knobel, et al. (2022a); Zare Sakhvidi, Yang, et al. (2022b); Zhang et al. (2021); Zhao et al. (2022); Zhao et al. (2021)
Green land cover 15 (33.3%) Batterham et al. (2022); Bolanis et al. (2024); Browning et al. (2022); Cao et al. (2023); Gao et al. (2024); Rahimi-Ardabili et al. (2021); Ricciardi et al. (2022); Rigolon et al. (2021); Rojas-Rueda et al. (2021); Wang et al. (2022); Ye et al. (2022); Zagnoli et al. (2022); Zare Sakhvidi, Yang, et al. (2022b); Zhang et al. (2021); Zhao et al. (2022)
Tree canopy 10 (22.2%) Ahmer et al. (2024); Bianconi et al. (2023); Bolanis et al. (2024); Gao et al. (2024); Patwary et al. (2024); Ricciardi et al. (2022); Wang et al. (2022); Yuan et al. (2021); Zagnoli et al. (2022); Zhao et al. (2022)
SAVI c 8 (17.8%) Aarthi et al. (2023); Bianconi et al. (2023); Patwary et al. (2024); Rahimi-Ardabili et al. (2021); Wang et al. (2022); Ye et al. (2022); Zare Sakhvidi, Knobel, et al. (2022a); Zhao et al. (2022)
EVI d 8 (17.8%) Cao et al. (2023); Ccami-Bernal et al. (2023); Patwary et al. (2024); Ricciardi et al. (2022); Wang et al. (2022); Ye et al. (2022); Y. D. Zhang et al. (2024b); Zhao et al. (2022)
Availability of greenspace 8 (17.8%) Browning et al. (2022); Li and Lange (2023); Masdor et al. (2023); Rojas-Rueda et al. (2021); Yuan et al. (2021); X. Zhang et al. (2022c); Y. D. Zhang et al. (2024b); Zhao et al. (2022)
Gardening 7 (15.6%) Briggs et al. (2022); Hsueh et al. (2022); Lampert et al. (2021); Rigolon et al. (2021); Tharrey and Darmon (2021); Tu (2022); Y. D. Zhang et al. (2024b)
Other a 37 (82.2%) Aarthi et al. (2023); Ahmer et al. (2024); Batterham et al. (2022); Bolanis et al. (2024); Browning et al. (2022); Cao et al. (2023); Ccami-Bernal et al. (2023); Chae et al. (2021); Chen et al. (2022); De la Fuente et al. (2021); Gao et al. (2024); Hsueh et al. (2022); Kang et al. (2022); Li and Lange (2023); Masdor et al. (2023); Patwary et al. (2024); Piva et al. (2024); Qiu et al. (2022); Rahimi-Ardabili et al. (2021); Ricciardi et al. (2022); Rigolon et al. (2021); Rojas-Rueda et al. (2021); Siah et al. (2023); Sivak et al. (2021); Wang et al. (2022); Wu et al. (2022); Ye et al. (2022); Yuan et al. (2021); Zagnoli et al. (2022); Zare Sakhvidi, Knobel, et al. (2022a); Zhang et al. (2021); X. Zhang et al. (2022c); Zhao et al. (2022); Zhao et al. (2021)

Other: proportion of greenspace, green exercise, frequency of greenspace exposure, viewing greenspace, density of greenspace, Light Detection and Ranging (LiDAR), greenspace exposure via Geographic Information System (GIS), land use, street view greenness, viewing greenspace photography/videography, Modified Soil Adjusted Vegetation Index (MSAVI), number of street trees, forest therapy, duration of greenspace exposure, forest bathing, Green View Index (GVI), greenspace ratio, number of green recreational facilities, Vegetative Continuous Field (VCF), greenness exposure by postcode, plant diversity, greenspace fragmentation, greenness satisfaction, greenspaces per inhabitant, greenspace area, quality of greenspace, intensity of exposure to nature, Leaf Area Index (LAI), total land area, type of greenspace, landscape elements, perceived quality and quantity of greenspace, activities in forest settings, loss of trees, biodiversity index, Moderate Resolution Imaging Spectroradiometer (MODIS), green patch area, green vision index, Vegetation Fractional Cover (VFC), green score, greening volunteerism, ecosystem count, nature meditation, nature-based programs, coordination of information of the environment (CORINE) land cover (CLC), undefined exposure to greenspace;

NDVI: Normalised Difference Vegetation Index;

SAVI: Soil-Adjusted Vegetation Index;

EVI: Enhanced Vegetation Index.

 Health outcomes

Health outcomes were sorted into the following categories: mental health and cognitive function, maternal health and birth outcomes, cardiovascular and metabolic outcomes, respiratory health and allergies, cancer, general health and QoL, and all-cause and cause-specific mortality. Health outcomes which did not align with these categories were classified as other health outcomes. Details of these categories and related ICD-10 codes are summarised in Table 4.

Health outcomes investigated in the included systematic reviews.
Health outcome category Health outcome ICD-10 code(s) a Review articles
Mental health and cognitive function Dementia A50.4, B22.0, F00-F03, F84.3, G31.0 Zagnoli et al. (2022); Zhao et al. (2022)
Overall mental health/psychological outcomes F00-F99 Batterham et al. (2022); Briggs et al. (2022); Hsueh et al. (2022); Kang et al. (2022); Lampert et al. (2021); Li and Lange (2023); Rahimi-Ardabili et al. (2021); Rojas-Rueda et al. (2021); Siah et al. (2023); Sivak et al. (2021); Tharrey and Darmon (2021); Tu (2022); Ye et al. (2022); Yi et al. (2022); Zhang et al. (2021)
Alzheimer’s dementia F00 Zagnoli et al. (2022); Zhao et al. (2021)
Vascular dementia F01 Zagnoli et al. (2022)
Non-Alzheimer’s dementia B22.0, F01-F03, F10.7, F11.7, F12.7, F13.7, F14.7, F15.7, F16.7, F17.7, F18.7, F19.7, F84.3, G31.0 Zagnoli et al. (2022); Zhao et al. (2021)
Mood, affect F06.3, F30-F39 Batterham et al. (2022); Briggs et al. (2022); Chen et al. (2022); Kang et al. (2022); Li and Lange (2023); Piva et al. (2024); Rahimi-Ardabili et al. (2021); Tharrey and Darmon (2021); Tu (2022); Yi et al. (2022)
Anxiety F06.4, F34.1, F40-F41, F51.5, F93.0-F93.2, Z63.7 Batterham et al. (2022); Briggs et al. (2022); Hsueh et al. (2022); Kang et al. (2022); Lampert et al. (2021); Rahimi-Ardabili et al. (2021); Siah et al. (2023); Tu (2022); Ye et al. (2022); Yi et al. (2022); Zhang et al. (2021); X. Zhang et al. (2022c)
Schizophrenia F20-F29 Batterham et al. (2022)
Depression F32-F33 Batterham et al. (2022); Briggs et al. (2022); Chen et al. (2022); Hsueh et al. (2022); Kang et al. (2022); Lampert et al. (2021); Piva et al. (2024); Rahimi-Ardabili et al. (2021); Rojas-Rueda et al. (2021); Siah et al. (2023); Sivak et al. (2021); Tu (2022); Ye et al. (2022); Yi et al. (2022); Zhang et al. (2021)
Stress F43, Z73 Briggs et al. (2022); Lampert et al. (2021); Li and Lange (2023); Piva et al. (2024); Rahimi-Ardabili et al. (2021); Siah et al. (2023); Sivak et al. (2021); Tharrey and Darmon (2021); Tu (2022); Ye et al. (2022)
Post traumatic stress disorder F43.1 Briggs et al. (2022); Hsueh et al. (2022); Lampert et al. (2021); Tu (2022)
Autism F84.0-F84.1 Ye et al. (2022)
Behavioural and psychological distress, emotional wellbeing, neurocognitive development F90-F98 Chen et al. (2022); Tharrey and Darmon (2021); Zare Sakhvidi, Knobel, et al. (2022a); Zhang et al. (2021)
Attention deficit hyperactivity disorder F90.0 Rahimi-Ardabili et al. (2021); Ye et al. (2022); Zare Sakhvidi, Knobel, et al. (2022a)
Alzheimer’s disease G30 Zagnoli et al. (2022)
Cognition R40-R44 Li and Lange (2023); Piva et al. (2024); Rahimi-Ardabili et al. (2021); Ricciardi et al. (2022); Rojas-Rueda et al. (2021); Siah et al. (2023); Tu (2022); Ye et al. (2022); Zagnoli et al. (2022); Zare Sakhvidi, Knobel, et al. (2022a); Zhao et al. (2022)
Suicidal ideation R45.8 Bolanis et al. (2024)
Self-harm X60-X84 Bolanis et al. (2024)
Neurochemical response Z00.4 Piva et al. (2024)
Burnout Z73.0 Briggs et al. (2022)
Maternal health and birth outcomes Overall birth and maternal health P00-P96, O00-O99 Browning et al. (2022)
Birthweight P07, P08 Ahmer et al. (2024); Rahimi-Ardabili et al. (2021)
Low birthweight P07.0, P07.1 Ahmer et al. (2024)
Preterm birth P07.2, P07.3 Ahmer et al. (2024); Rigolon et al. (2021)
Small for gestational age P05.1 Ahmer et al. (2024)
Miscarriage O03 Rahimi-Ardabili et al. (2021)
Cardiovascular and metabolic health Diabetes E10-E14, E23.2, O24 Aarthi et al. (2023); Browning et al. (2022); Ccami-Bernal et al. (2023); De la Fuente et al. (2021); Rahimi-Ardabili et al. (2021); Rigolon et al. (2021)
Blood glucose level E10-E16, E23.2, O24, R73 Aarthi et al. (2023); Siah et al. (2023)
Metabolic health E70-E90 Ye et al. (2022)
Cholesterol E78 Siah et al. (2023)
Metabolic syndrome E88.8 Patwary et al. (2024); Rahimi-Ardabili et al. (2021)
Cardiovascular health I00-I99 Bianconi et al. (2023); Browning et al. (2022); Rahimi-Ardabili et al. (2021); Rigolon et al. (2021); Rojas-Rueda et al. (2021); Siah et al. (2023); Ye et al. (2022); Yuan et al. (2021)
Heart failure I09.0, I09.9 I11.0, I13.0, I43.8, I50, I97.1 Bianconi et al. (2023)
Hypertension I10-I15 Rahimi-Ardabili et al. (2021); Zhao et al. (2022)
Ischaemic heart disease I20-I25 Bianconi et al. (2023); Yuan et al. (2021)
Acute myocardial infarction I21-I22 Bianconi et al. (2023); Yuan et al. (2021)
Coronary heart disease I25.9 Bianconi et al. (2023); Rahimi-Ardabili et al. (2021); Yuan et al. (2021)
Stroke I64 Bianconi et al. (2023); Rahimi-Ardabili et al. (2021); Yuan et al. (2021)
Heart rate R00 Piva et al. (2024); Siah et al. (2023); Yi et al. (2022)
Blood pressure Z01.3 Piva et al. (2024); Qiu et al. (2022); Rahimi-Ardabili et al. (2021); Siah et al. (2023); Sivak et al. (2021); Tharrey and Darmon (2021); Ye et al. (2022); Yi et al. (2022); Zhao et al. (2022)
Respiratory health and allergies Acute rhino conjunctivitis H10 Wang et al. (2022)
Respiratory diseases J00-J99 Browning et al. (2022); Rahimi-Ardabili et al. (2021); Rojas-Rueda et al. (2021)
Rhinitis J00, J30.0, J31.0 Rahimi-Ardabili et al. (2021); Ye et al. (2022)
Pneumonia J14-J18, J69, J84.9, P23, P24.9 Ye et al. (2022)
Bronchitis J20, J40-J42, J44, J45.0, J45.9 Ye et al. (2022)
Allergic rhinitis J30, J45.0 Cao et al. (2023); Tang et al. (2023); Wang et al. (2022); Wu et al. (2022); Ye et al. (2022)
Chronic obstructive pulmonary disorder J44 Gao et al. (2024); Rahimi-Ardabili et al. (2021); Tang et al. (2023)
Asthma J45-J46 Rahimi-Ardabili et al. (2021); Rigolon et al. (2021); Tang et al. (2023); Wang et al. (2022); Wu et al. (2022); Ye et al. (2022)
Eczema & dermatitis L20-L30 Rahimi-Ardabili et al. (2021); Rigolon et al. (2021); Wang et al. (2022); Ye et al. (2022)
Food allergies L27.2, L23.6, T78.0-T78.1 Wang et al. (2022); Ye et al. (2022)
Lung function Z00.8 Piva et al. (2024); Tharrey and Darmon (2021); Ye et al. (2022)
Cancer All-site cancer C00-D48 Browning et al. (2022); Li et al. (2023); Zare Sakhvidi, Yang, et al. (2022b)
Non-melanoma cancer C00-C41, C45-D02, D04-D09 Zare Sakhvidi, Yang, et al. (2022b)
Mouth cancer C00-C09, D00.0 Li et al. (2023); Zare Sakhvidi, Yang, et al. (2022b)
Throat cancer C09-C13, C14.0, D00.0 Zare Sakhvidi, Yang, et al. (2022b)
Colorectal cancer C19, C78.5, D01.1 Li et al. (2023); Masdor et al. (2023); Zare Sakhvidi, Yang, et al. (2022b)
Lung cancer C34, C78.0, D02.2 Li et al. (2023); Rahimi-Ardabili et al. (2021); Tang et al. (2023); Zare Sakhvidi, Yang, et al. (2022b)
Skin cancer C43-C44, C79.2, D03-D04 Li et al. (2023); Zare Sakhvidi, Yang, et al. (2022b)
Breast cancer C50, D24, D05 Li et al. (2023); Zare Sakhvidi, Yang, et al. (2022b)
Prostate cancer C61, D07.5 Li et al. (2023); Rigolon et al. (2021); Zare Sakhvidi, Yang, et al. (2022b)
Bladder cancer C67, C79.1, D09.0 Li et al. (2023)
Brain tumour C71, D33, C79.3, D43.0-D43.2 Zare Sakhvidi, Yang, et al. (2022b)
Leukaemia C90.1, C91-C95, D47.1, D47.5 Ye et al. (2022)
General health and quality of life Overall/general health/physiological outcomes A00-Y98 Lampert et al. (2021); Li and Lange (2023); Rahimi-Ardabili et al. (2021); Rigolon et al. (2021); Rojas-Rueda et al. (2021); Siah et al. (2023); Sivak et al. (2021); Tharrey and Darmon (2021); Ye et al. (2022); Yi et al. (2022)
Chronic diseases A00-Q99 Lampert et al. (2021); Rahimi-Ardabili et al. (2021)
Pain G50.1, H57.1, I20.9, K14.6, M25.5, M54.5, M54.6, M79.6, N94, R07, R10, R14, R30, R51-R52 Siah et al. (2023)
Fatigue R53 Rahimi-Ardabili et al. (2021); Siah et al. (2023)
Quality of life NA b Briggs et al. (2022); Lampert et al. (2021); Piva et al. (2024); Rahimi-Ardabili et al. (2021); Rojas-Rueda et al. (2021); Siah et al. (2023); Ye et al. (2022); Yi et al. (2022)
Life expectancy NA Rojas-Rueda et al. (2021)
All-cause and cause specific mortality All-cause mortality A00-V98 Browning et al. (2022); Rahimi-Ardabili et al. (2021); Rigolon et al. (2021); Yuan et al. (2021)
Non-accidental mortality A00-T98, X60-Y09, Y35-Y89 Yuan et al. (2021)
Infectious & parasitic disease mortality A00-B99 Yuan et al. (2021)
Dementia mortality A50.4, B22.0, F00-F03, F84.3, G31.0 Zagnoli et al. (2022)
All-site cancer mortality C00-D48 Li et al. (2023); Yuan et al. (2021); Zare Sakhvidi, Yang, et al. (2022b)
Oesophageal cancer mortality C15, D00.1 Zare Sakhvidi, Yang, et al. (2022b)
Colorectal cancer mortality C19, D01.1 Masdor et al. (2023)
Lung cancer mortality C34, C78.0 Li et al. (2023); Tang et al. (2023); Zare Sakhvidi, Yang, et al. (2022b)
Prostate cancer mortality C61 Masdor et al. (2023); Zare Sakhvidi, Yang, et al. (2022b)
Neurogenerative disease mortality F00, G10-G37, G60-G99, R27.0, R27.8 Yuan et al. (2021); Zagnoli et al. (2022)
Circulatory mortality I00-I99 Yuan et al. (2021)
Ischaemic heart disease mortality I20-I25 Bianconi et al. (2023); Yuan et al. (2021)
Acute myocardial infarction I21-I22 Yuan et al. (2021)
Cardiovascular disease mortality I51.6 Bianconi et al. (2023); Yuan et al. (2021)
Stroke mortality I64 Bianconi et al. (2023); Yuan et al. (2021)
Respiratory disease mortality J00-J99 Yuan et al. (2021)
COPD mortality J44 Tang et al. (2023)
Infant/neonatal mortality P00-P04, P07-P08, P95 Rigolon et al. (2021); Rojas-Rueda et al. (2021)
Suicide mortality T76, Z60-X84 Bolanis et al. (2024)
Other health outcomes Foodborne pathogens A02, A04.7, A05, A32 Sivak et al. (2021)
Dysentery A07.9, A09.0 Rahimi-Ardabili et al. (2021)
Tuberculosis A15-A19 Rahimi-Ardabili et al. (2021)
West Nile virus A92.3 Sivak et al. (2021)
Dengue A97 Rahimi-Ardabili et al. (2021)
Dermatophytosis B35 Sivak et al. (2021)
Malaria B50-B54 Rahimi-Ardabili et al. (2021)
Intestinal parasites B65-B83 Ye et al. (2022)
Toxocariasis B83.0 Sivak et al. (2021)
Cortisol level E24, E27.1-E27.2, E27.8, E35.1 Piva et al. (2024); Qiu et al. (2022); Siah et al. (2023)
Parkinson’s disease G20 Zagnoli et al. (2022)
Multiple sclerosis G35 Zagnoli et al. (2022)
Myopia H44.2, H52.1 Ye et al. (2022)
Astigmatism H52.2 Ye et al. (2022)
Irritable bowel disease K51, K63.8 Ye et al. (2022)
Heavy metal contamination R82.6, R78.7, T56, X49, Y19 Sivak et al. (2021)
Microbiota (gut, skin, oral, nasal) Z01.8 Y. D. Zhang et al. (2024b)
Immune markers Z01.8 Chae et al. (2021)

ICD-10: International Classification of Diseases;

NA: Not applicable.

Mental health and cognitive function

A total of 24 systematic reviews examined the association between greenspace exposure and mental health or cognitive function. Overall mental health and well-being were investigated in 11 systematic reviews ( Batterham et al., 2022; Chen et al., 2022; Hsueh et al., 2022; Kang et al., 2022; Lampert et al., 2021; Rahimi-Ardabili et al., 2021; Tharrey & Darmon, 2021; Tu, 2022; Ye et al., 2022; Zare Sakhvidi, Knobel, et al., 2022a; Zhang et al., 2021). Greenspace interventions, such as greening volunteerism ( Chen et al., 2022), collective gardening ( Tharrey & Darmon, 2021), community gardening ( Lampert et al., 2021), and horticultural therapy ( Tu, 2022), showed positive effects on mental health and wellbeing. Forest-based interventions (encompassing forest therapy, forest bathing, forest videos, and forest-walking programs) ( Kang et al., 2022) and overall greenspace exposure (encompassing area of greenspace, area of parks, and green coverage ratio) ( Rahimi-Ardabili et al., 2021) were linked to better mental health, including in children ( Ye et al., 2022). Mediating factors between greenspace and mental health, such as air quality, perceived stress, and physical activity, were also identified ( Zhang et al., 2021). However, some reviews reported mixed evidence on the relationship between greenspace exposure and overall mental health and wellbeing ( Batterham et al., 2022; Hsueh et al., 2022; Zare Sakhvidi, Knobel, et al., 2022a).

Depression was assessed in eight reviews ( Batterham et al., 2022; Briggs et al., 2022; Kang et al., 2022; Piva et al., 2024; Siah et al., 2023; Sivak et al., 2021; Yi et al., 2022; Y. Zhang et al., 2024a). Meta-analyses demonstrated protective effects of group-based gardening ( Briggs et al., 2022) and NDVI ( Y. Zhang et al., 2024a), though heterogeneity was a limitation. Protective effects were reported for forest-based interventions ( Kang et al., 2022; Piva et al., 2024; Siah et al., 2023; Yi et al., 2022), but narrative synthesis yielded mixed results for general greenspace ( Batterham et al., 2022; Briggs et al., 2022) and vacant lots ( Sivak et al., 2021).

Anxiety outcomes appeared in five reviews ( Briggs et al., 2022; Kang et al., 2022; Siah et al., 2023; X. Zhang et al., 2022c; Y. Zhang et al., 2024a). Increased greenspace exposure ( X. Zhang et al., 2022c) and forest-based interventions ( Kang et al., 2022; Siah et al., 2023) were associated with reduced anxiety levels. Increased greenspace, specifically NDVI, was also associated with lower odds of anxiety ( Y. Zhang et al., 2024a). However, evidence for group-based gardening was inconsistent in narrative synthesis and no association was found in the meta-analysis ( Briggs et al., 2022).

Cognitive function was assessed in six reviews ( Rahimi-Ardabili et al., 2021; Ricciardi et al., 2022; Rojas-Rueda et al., 2021; Zagnoli et al., 2022; Y. Zhang et al., 2024a; Zhao et al., 2021). Meta-analyses concluded that higher greenspace was associated with a reduced risk of combined cognitive impairment and dementia ( Zhao et al., 2021) and increased NDVI was associated with a decreased odds of dementia ( Y. Zhang et al., 2024a). Conversely, Zagnoli et al. (2022) reported that increased greenspace was associated with an increased risk of cognitive impairment, though this finding was limited by the small number of original studies. Inconsistent ( Ricciardi et al., 2022), insufficient ( Rojas-Rueda et al., 2021), or no evidence ( Rahimi-Ardabili et al., 2021) was noted in other reviews.

Other mental and cognitive health outcomes were also explored ( Bolanis et al., 2024; Briggs et al., 2022; Kang et al., 2022; Li & Lange, 2023; Piva et al., 2024; Rojas-Rueda et al., 2021; Siah et al., 2023; Sivak et al., 2021; Zare Sakhvidi, Knobel, et al., 2022a; Y. Zhang et al., 2024a). Increased NDVI within 400m (OR [odds ratio] = 0.59, 95%CI: 0.35, 0.99) and 1500m (OR = 0.52, 95%CI: 0.30, 0.91) of residences was associated with reduced odds of non-psychotic mental disorders ( Rojas-Rueda et al., 2021). Higher NDVI was also linked to lower odds of schizophrenia and attention deficit hyperactivity disorder (ADHD) ( Y. Zhang et al., 2024a). Green vacant lots ( Sivak et al., 2021) and passive or light physical activity in urban greenspaces ( Li & Lange, 2023) were associated with reduced psychological stress while forest-based interventions were associated with emotions ( Kang et al., 2022) and mood ( Piva et al., 2024; Siah et al., 2023). Protective associations were suggested for self-harm and suicidal ideation ( Bolanis et al., 2024), but evidence was mixed for conduct problems and internalising/externalising disorders ( Zare Sakhvidi, Knobel, et al., 2022a). Overall, evidence on the association between greenspace and mental health and cognitive outcomes was mostly beneficial.

Maternal health and birth outcomes

Only five reviews investigated maternal health and birth outcomes ( Ahmer et al., 2024; Browning et al., 2022; Rahimi-Ardabili et al., 2021; Rigolon et al., 2021; Rojas-Rueda et al., 2021). The evidence on birth outcomes was notably limited, as highlighted by Rojas-Rueda et al. (2021) and Rahimi-Ardabili et al. (2021) who only found one original study each. A meta-analysis showed that increased residential greenspace within a 250m radius was associated with higher birthweights (BW) and reduced odds of low birthweight (LBW) ( Ahmer et al., 2024). Narrative synthesis also suggested that residential greenspace may protect against preterm birth (PTB) and small-for-gestational-age (SGA) births ( Ahmer et al., 2024). Urban settings appeared to enhance the protective effects of greenspace on birth outcomes ( Browning et al., 2022). There was no evidence that socioeconomic status (SES) was an effect modifier ( Rigolon et al., 2021). Although there were limited systematic reviews on maternal health and birth outcomes, evidence showed protective associations.

Cardiovascular and metabolic outcomes

Cardiovascular and metabolic outcomes were examined in 17 reviews ( Aarthi et al., 2023; Bianconi et al., 2023; Browning et al., 2022; Ccami-Bernal et al., 2023; De la Fuente et al., 2021; Patwary et al., 2024; Piva et al., 2024; Qiu et al., 2022; Rahimi-Ardabili et al., 2021; Rigolon et al., 2021; Rojas-Rueda et al., 2021; Siah et al., 2023; Tharrey & Darmon, 2021; Ye et al., 2022; Yi et al., 2022; Yuan et al., 2021; Zhao et al., 2022). Higher greenspace exposure was associated with a lower incidence of diabetes mellitus ( Ccami-Bernal et al., 2023; Rahimi-Ardabili et al., 2021) and reduced risk of type 2 diabetes mellitus (T2DM) ( De la Fuente et al., 2021). However, Aarthi et al. (2023) reported inconsistent evidence for this association. Greenspace exposure also showed inverse relationships with CVD, ischaemic heart disease (IHD), acute myocardial infarction (AMI), and stroke morbidity, except when greenspace was measured via tree canopy exposure ( Bianconi et al., 2023).

Forest-based interventions were beneficially associated with systolic blood pressure (SBP) and diastolic blood pressure (DBP) in some reviews ( Piva et al., 2024; Qiu et al., 2022; Zhao et al., 2022) while others found no notable effects ( Siah et al., 2023; Yi et al., 2022). Zhao et al. (2022) reported associations between NDVI and blood pressure (BP)/hypertension at buffers of 100m (OR = 0.91, 95%CI: 0.86, 0.97), 250-300m (OR = 0.98, 95%CI: 0.97, 0.99), 500m (OR = 0.94, 95%CI: 0.94, 0.95), and 1000m (OR = 0.98, 95%CI: 0.97, 0.99), but other systematic reviews reported inconsistent ( Ye et al., 2022) or inadequate evidence ( Tharrey & Darmon, 2021) on BP outcomes.

An increase in NDVI within a 500m buffer was associated with lower odds of metabolic syndrome (MetS) ( Patwary et al., 2024). Heart rate (HR) showed minimal ( Piva et al., 2024) or no change ( Rahimi-Ardabili et al., 2021; Siah et al., 2023) in response to greenspace exposure. Greenspace exposure was linked to reduced CVD risk, but evidence on stroke and myocardial infarction remained inconsistent ( Yuan et al., 2021). Evidence suggested that beneficial effects of greenspace on cardiovascular outcomes were stronger in urban areas than in less urban areas ( Browning et al., 2022). Benefits were also larger for individuals with lower SES compared to those with higher SES ( Rigolon et al., 2021). While some systematic reviews on cardiovascular and metabolic outcomes reported inconsistent associations, most found beneficial results.

Respiratory health and allergies

A total of 12 reviews addressed respiratory health and allergies ( Browning et al., 2022; Cao et al., 2023; Gao et al., 2024; Piva et al., 2024; Rahimi-Ardabili et al., 2021; Rigolon et al., 2021; Rojas-Rueda et al., 2021; Tang et al., 2023; Tharrey & Darmon, 2021; Wang et al., 2022; Wu et al., 2022; Ye et al., 2022). Higher residential greenness at birth was associated with lower odds of allergic rhinitis (AR) in childhood (OR = 0.83, 95%CI: 0.72, 0.96) ( Wang et al., 2022) and lower odds of current asthma when measured within a 500m buffer ( Wang et al., 2022). Similarly, a 0.1-unit increment in NDVI was associated with a reduced risk of asthma incidence (RR [relative risk] = 0.92, 95%CI: 0.85, 0.98) ( Tang et al., 2023). However, other reviews found no associations between greenspace exposure and AR ( Cao et al., 2023; Wu et al., 2022; Ye et al., 2022) or asthma ( Wu et al., 2022; Ye et al., 2022).

Greenspace exposure demonstrated beneficial associations with lung function in children ( Ye et al., 2022) and in individuals with chronic obstructive pulmonary disorder (COPD) ( Gao et al., 2024). Mixed findings were reported for overall respiratory outcomes ( Rahimi-Ardabili et al., 2021), atopic dermatitis ( Wang et al., 2022), and food allergies ( Wang et al., 2022). Inadequate evidence was noted in other reviews ( Piva et al., 2024; Rojas-Rueda et al., 2021; Tharrey & Darmon, 2021). Urbanicity and SES did not modify the association between greenspace and respiratory health and atopic diseases ( Browning et al., 2022; Rigolon et al., 2021). Altogether, evidence on respiratory health and allergies was mixed.

Cancer

The association between greenspace exposure and cancer was explored in eight reviews ( Browning et al., 2022; Li et al., 2023; Masdor et al., 2023; Rahimi-Ardabili et al., 2021; Rigolon et al., 2021; Tang et al., 2023; Ye et al., 2022; Zare Sakhvidi, Yang, et al., 2022b). Tang et al. (2023) reported that a 0.1-unit increment in NDVI was associated with a reduced risk of lung cancer incidence, although high heterogeneity was observed across studies.

Insufficient evidence was found for an association between greenspace exposure and colorectal cancer ( Li et al., 2023; Masdor et al., 2023), lung cancer ( Rahimi-Ardabili et al., 2021; Zare Sakhvidi, Yang, et al., 2022b), breast cancer ( Zare Sakhvidi, Yang, et al., 2022b), prostate cancer ( Zare Sakhvidi, Yang, et al., 2022b), and all-site cancer ( Li et al., 2023) based on both quantitative and qualitative syntheses. Li et al. (2023) highlighted inconsistent findings for prostate, lung, breast, bladder, and skin cancers, further underscoring the need for more robust and comprehensive research to clarify the role of greenspace exposure in cancer outcomes. Other reviews were limited by insufficient evidence ( Browning et al., 2022; Rigolon et al., 2021; Ye et al., 2022). For the most part, mixed findings or no association were reported among systematic reviews on the association between greenspace and cancer.

General health and quality of life

General health and QoL were assessed in 12 reviews ( Briggs et al., 2022; Lampert et al., 2021; Li & Lange, 2023; Piva et al., 2024; Rahimi-Ardabili et al., 2021; Rigolon et al., 2021; Rojas-Rueda et al., 2021; Siah et al., 2023; Sivak et al., 2021; Tharrey & Darmon, 2021; Ye et al., 2022; Yi et al., 2022). Forest bathing was linked to enhanced QoL ( Siah et al., 2023), while forest therapy demonstrated associations with improved physiological parameters ( Piva et al., 2024). Visits to parks with a companion (PR [prevalence ratio] = 1.12, 95% CI: 1.01–1.25) and the condition of trees in parks (PR = 1.20, 95% CI: 1.07–1.34) were also positively associated with QoL ( Rojas-Rueda et al., 2021). However, evidence regarding forest therapy and QoL was mixed ( Yi et al., 2022).

Li and Lange (2023) found that passive exposure to greenspace or light physical activity in green settings reduced physical stress. Gardening interventions were associated with better physical health ( Lampert et al., 2021) and wellbeing (SMD [standardised mean difference] = 0.37, 95% CI: 0.01, 0.73) ( Briggs et al., 2022). However, evidence on gardening and physiological health was mixed ( Tharrey & Darmon, 2021), and no association was observed with health-related QoL ( Briggs et al., 2022). In narrative synthesis, greenspace exposure was linked to improved general and physical health but not frailty, disability, or chronic diseases ( Rahimi-Ardabili et al., 2021). Rigolon et al. (2021) identified SES as an important modifier in the relationship between greenspace and general health. Evidence on the risk of physical injury ( Sivak et al., 2021), hospital admission ( Ye et al., 2022), and sleep quality/sufficiency ( Ye et al., 2022) was limited. Overall, systematic reviews on general health and QoL showed a protective effect of greenspace exposure.

All-cause and cause-specific mortality

All-cause and cause-specific mortality was examined in 12 reviews ( Bianconi et al., 2023; Bolanis et al., 2024; Browning et al., 2022; Li et al., 2023; Masdor et al., 2023; Rahimi-Ardabili et al., 2021; Rigolon et al., 2021; Rojas-Rueda et al., 2021; Tang et al., 2023; Yuan et al., 2021; Zagnoli et al., 2022; Zare Sakhvidi, Yang, et al., 2022b). Incremental increases in NDVI were associated with reduced mortality risks from lung cancer ( Li et al., 2023), prostate cancer ( Li et al., 2023), and COPD ( Tang et al., 2023). Bianconi et al. (2023) reported that urban greenspace was associated with a reduced risk of CVD, IHD, and cerebrovascular disease (CBVD) mortality. Greenspace also showed a favourable association with childhood mortality ( Rojas-Rueda et al., 2021). The protective association between greenspace and suicide mortality was stronger in females than males ( Bolanis et al., 2024) and the association with all-cause mortality was stronger in urban than less urban areas ( Browning et al., 2022).

Inconsistent or insufficient evidence was reported for associations with mortality due to prostate cancer ( Li et al., 2023; Masdor et al., 2023), lung cancer ( Tang et al., 2023), oesophageal cancer ( Zare Sakhvidi, Yang, et al., 2022b), all-cause cancer ( Yuan et al., 2021), stroke ( Yuan et al., 2021), dementia ( Zagnoli et al., 2022), and all-cause mortality ( Rahimi-Ardabili et al., 2021). Overall, evidence on all-cause and cause-specific mortality was inconsistent.

Other health outcomes

A total of seven reviews explored health outcomes which did not fit the above categories ( Chae et al., 2021; Piva et al., 2024; Qiu et al., 2022; Rahimi-Ardabili et al., 2021; Sivak et al., 2021; Ye et al., 2022; Y. D. Zhang et al., 2024b). Forest therapy was associated with an increase in natural killer (NK) cell counts and activity, as well as changes in cytotoxic effector molecules and proinflammatory cytokines ( Chae et al., 2021). It also decreased serum cortisol concentration (SCC) in participants who underwent forest therapy compared to urban controls (MD [mean difference] = -0.07, 95%CI: -1.10, -0.04), though high heterogeneity (I 2 = 83.9%) was noted ( Qiu et al., 2022). Walking in forests was associated with improved neurochemical markers, including reductions in interleukin-6 (IL-6) and tumor necrosis factor alpha (TNF-α) levels ( Piva et al., 2024).

Rahimi-Ardabili et al. (2021) linked greenspace exposure to a lower incidence of dysentery but an increased incidence of malaria, dengue fever, and tuberculosis in China. However, these conclusions were constrained by limited original studies. Protective trends were suggested for infections, myopia, and astigmatism, though insufficient evidence precluded definitive conclusions ( Ye et al., 2022).

Vacant lots with greenspace were associated with elevated blood lead levels in children and increased rates of mosquitos carrying West Nile Virus ( Sivak et al., 2021). Connections between vacant lots and dermatophytosis and toxocariasis were also indicated ( Sivak et al., 2021). Y. D. Zhang et al. (2024b) reported that greenspace exposure was positively associated with the diversity and composition of gut and skin microbiota. It was also linked to an increase in probiotics and a reduction in pathogens, but evidence for nasal, ocular, and oral microbiota was inconclusive ( Y. D. Zhang et al., 2024b).

 Quantitative data summary

While this review did not conduct quantitative synthesis, Table 5 provides a comprehensive list of the quantitative measures reported across the systematic reviews. Effect estimates based on more than one original study were listed in this table.

Listing of quantitative data from included reviews' meta-analyses.
Outcome Review article No. & type of original studies included Exposure measure Effect measure Estimate (95%CI a ) from meta-analysis
Mental health and cognitive function
Dementia Zagnoli et al. (2022) 6 (2 cross-sectional, 2 cohort, 1 combined cohort case-control) NDVI b RR c (highest vs lowest greenness categories, 1-unit increases) 0.98 (0.90, 1.06)
6 (2 cross-sectional, 2 cohort, 1 case-control, 1 ecological Land use/land change 0.99 (0.93, 1.05)
Y. Zhang et al. (2024a) 8 (cross-sectional) Any greenspace exposure OR d (dose-response not reported) 0.95 (0.93, 0.96)
5 (cohort) NDVI 0.95 (0.94, 0.96)
3 (cohort) Area of greenspace 0.93 (0.84, 1.03)
Dementia/cognitive impairment Zhao et al. (2021) 8 (4 cohort, 2 cross-sectional, 2 case-control) Residential greenness OR (dose-response not reported) 0.97 (0.95, 1.00)
4 (cohort) 0.96 (0.95, 0.98)
Cognitive impairment Zagnoli et al. (2022) 2 (cross-sectional) Land use/land change RR (highest vs lowest greenness categories) 1.47 (1.22, 1.76)
Depression Briggs et al. (2022) 8 (RCT) Gardening SMD e (intervention vs control) -0.43 (-0.79, -0.06)
Kang et al. (2022) 5 (RCT) Forest therapy SMD (pre- vs post-intervention) 1.36 (0.55, 2.17)
Siah et al. (2023) 10 (5 quasi-experimental, 4 RCT, 1 crossover design) Forest bathing SMD (intervention vs control) -0.67 (-0.99, -0.35)
Yi et al. (2022) 4 (RCT) Forest therapy SMD (intervention vs control) -1.46 (-2.80, -0.12)
Y. Zhang et al. (2024a) 37 (29 cross-sectional, 8 cohort) Any greenspace exposure Any greenspace exposure OR (dose-response not reported) 0.89 (0.86, 0.93)
35 (28 cross-sectional, 7 cohort) 0.96 (0.94, 0.97)
5 (cross-sectional) View of greenery 0.89 (0.74, 1.06)
16 (12 cross-sectional, 4 cohort) NDVI 0.95 (0.91, 0.98)
19 (16 cross-sectional, 3 cohort) Area of greenspace 0.97 (0.94, 1.01)
3 (cross-sectional) Green space accessibility 0.85 (0.67, 1.08)
3 (2 cross-sectional, 1 cohort) Parks 1.00 (0.99, 1.00)
4 (2 cohort, 2 cross-sectional) Other greenspace exposures f 0.88 (0.83, 0.94)
Anxiety Briggs et al. (2022) 5 (RCT) Gardening SMD (intervention vs control) -0.42 (-1.00, 0.16)
Kang et al. (2022) 5 (RCT) Forest therapy SMD (pre- vs post-intervention) 0.88 (0.18, 1.58)
Siah et al. (2023) 7 (4 RCT, 2 quasi-experimental, 1 crossover design) Forest bathing SMD (intervention vs control) -0.84 (-1.42, -0.25)
X. Zhang et al. (2022c) 24 (12 randomised crossover, 7 non-randomised crossover, 3 randomised parallel group, 1 2x2 factorial design, 1single group crossover) Natural environment SMD (pre- vs post-intervention) -1.28 (-1.65, -0.92)
Y. Zhang et al. (2024a) 14 (11 cross-sectional, 3 cohort) Any greenspace exposure OR (dose-response not reported) 0.94 (0.92, 0.96)
9 (7 cross-sectional, 2 cohort) Any greenspace exposure 0.94 (0.91, 0.97)
3 (cross-sectional) View of greenery 0.84 (0.78, 0.91)
2 (cross-sectional) Area of greenspace 0.98 (0.93, 1.03)
3 (2 cross-sectional, 1 cohort) NDVI 0.95 (0.92, 0.98)
2 (1 cohort, 1 cross-sectional) Other greenspace exposures g 0.93 (0.85, 1.03)
Stress Briggs et al. (2022) 3 (RCT) Gardening SMD (intervention vs control) -0.17 (-0.68, 0.35)
ADHD h Y. Zhang et al. (2024a) 5 (3 cohort, 2 cross-sectional) Any greenspace exposure OR (dose-response not reported) 0.89 (0.86, 0.92)
Positive emotions Kang et al. (2022) 4 (RCT) Forest therapy SMD (pre- vs post-intervention) 0.91 (0.34, 1.47)
Negative emotions Kang et al. (2022) 3 (RCT) Forest therapy SMD (pre- vs post-intervention) 1.37(0.81, 1.93)
Psychiatric disorders Y. Zhang et al. (2024a) 60 (32 cross-sectional, 28 cohort) Any greenspace exposure OR (dose not reported) 0.91 (0.89, 0.92)
Schizophrenia Y. Zhang et al. (2024a) 7 (cohort) Any greenspace exposure OR (dose-response not reported) 0.74 (0.67, 0.82)
7 (cohort) Any greenspace exposure 0.75 (0.69, 0.81)
5 (cohort) NDVI 0.72 (0.64, 0.82)
4 (cohort) Area of greenspace 0.77 (0.70, 0.85)
Overall mental health Kang et al. (2022) 6 (RCT) Forest therapy SMD (pre- vs post-intervention) 1.25 (0.93, 1.57)
Tu (2022) 18 (RCT) Horticultural therapy Hedge’s g (intervention vs control) 0.55 (0.38, 0.72)
Maternal health and birth outcomes
Birthweight Ahmer et al. (2024) 8 (7 cross-sectional, 1 case-control) NDVI 250m buffer beta standardised regression coefficient (per 0.1-unit change) 8.95 (1.63, 16.27)
8 (7 cross-sectional, 1 case-control) NDVI 250m buffer 9.86 (1.91, 17.81)
11 (9 cross-sectional, 1 case-control, 1 cohort) NDVI 500m buffer 12.83 (4.69, 20.97)
13 (11 cross-sectional, 1 case-control, 1 cohort) NDVI 250m & 500m buffer 11.18 (5.69, 16.67)
Low birth weight Ahmer et al. (2024) 6 (5 cross-sectional, 1 cohort) NDVI 250m buffer OR (per 0.1-unit change) 0.97 (0.96, 0.98)
Cardiovascular and metabolic outcomes
Systolic blood pressure Qiu et al. (2022) 20 (10 parallel RCT, 8 crossover RCT, 2 non-RCT) Forest therapy MD i (change in intervention vs control) -3.44 (-5.74, -1.14)
Siah et al. (2023) 13 (9 RCT, 2 quasi-experimental, 1 non-randomised crossover, 1 crossover) Forest bathing MD (intervention vs control) -1.66 (-4.30, 0.97)
Yi et al. (2022) 6 (RCT) Forest therapy MD (intervention vs control) -0.24 (-2.70, 2.23)
Zhao et al. (2022) 5 (4 cross-sectional, 1 cohort) NDVI 500m buffer beta standardised regression coefficient (per 0.1-unit increase) -0.77 (-1.23, -0.32)
Diastolic blood pressure Qiu et al. (2022) 21 (10 parallel RCT, 9 crossover RCT, 2 non-RCT) Forest therapy MD (change in intervention vs control) -3.07 (-5.59, -0.54)
Siah et al. (2023) 13 (9 RCT, 2 quasi-experimental, 1 non-randomised crossover, 1 crossover) Forest bathing MD (intervention vs control) -3.09 (-7.52, 1.34)
Yi et al. (2022) 5 (RCT) Forest therapy (post-intervention vs control) MD (intervention vs control) 0.94 (-3.20, 5.07)
Zhao et al. (2022) 5 (4 cross-sectional, 1 cohort) NDVI 500m buffer beta standardised regression coefficient (per 0.1-unit increase) -0.32 (-0.57, -0.07)
Blood pressure/hypertension Zhao et al. (2022) 3 (cross-sectional) NDVI 100m buffer OR (per 0.1-unit increase) 0.91 (0.86, 0.97)
4 (cross-sectional) NDVI 250/300m buffer 0.98 (0.97, 0.99)
8 (7 cross-sectional, 1 case-control) NDVI 500m buffer 0.94 (0.93, 0.95)
4 (cross-sectional) NDVI 1000m buffer 0.98 (0.97, 0.99)
4 (3 cross-sectional, 1 longitudinal) Proportion of greenspace OR (per 1% increase) 0.99 (0.99, 1.00)
3 (cross-sectional) Distance to greenspace OR (per 500m difference) 1.03 (0.96, 1.10)
Heart rate Siah et al. (2023) 5 (3 RCT, 1 quasi-experimental, 1 crossover) Forest bathing MD (intervention vs control) -0.42 (-3.32, 2.49)
Metabolic syndrome Patwary et al. (2024) 4 (3 cohort, 1 cross-sectional) NDVI 500m buffer OR (dose-response not reported) 0.90 (0.87, 0.93)
Respiratory health and allergies
Asthma Tang et al. (2023) 9 (5 cross-sectional, 2 case-control, 2 cohort) NDVI 200-300m buffer RR (per 0.1-unit increase) 0.92 (0.86, 0.98)
9 (6 cross-sectional, 2 case-control, 1 cohort) NDVI 500m buffer 0.93 (0.85, 1.01)
8 (5 cross-sectional, 1 case-control, 1 cohort, 1 longitudinal) NDVI 1000m buffer 0.87 (0.81, 0.93)
Ye et al. (2022) 10 (5 cohort, 5 cross-sectional) NDVI OR (per 0.1-unit increase) 0.94 (0.84, 1.05)
Current asthma Wang et al. (2022) 3 (cross-sectional) NDVI 100m buffer OR (per IQR increase) 1.02 (0.90, 1.14)
3 (cross-sectional) NDVI 100-500m buffer 0.94 (0.78, 1.13)
4 (cross-sectional) NDVI 500m buffer 0.88 (0.78, 0.99)
4 (cross-sectional) NDVI 1000m buffer 0.94 (0.86, 1.02)
4 (cross-sectional) NDVI 100-1000m buffer 0.94 (0.88, 1.00)
Wu et al. (2022) 3 (2 cross-sectional, 1 case-control) NDVI >0-100m buffer OR (per 0.1-unit increase) 0.98 (0.90, 1.07)
4 (2 cross-sectional, 1 case-contol, 1 cohort) NDVI >100-300m buffer 0.99 (0.91, 1.07)
4 (2 cross-sectional, 1 case-contol, 1 cohort) NDVI >300-500m buffer 1.00 (0.91, 1.09)
4 (2 cross-sectional, 1 case-contol, 1 cohort) NDVI >500-1000m buffer 0.98 (0.90, 1.08)
Ever had asthma Wu et al. (2022) 4 (2 case-control, 2 cross-sectional) NDVI >0-100m buffer OR (per 0.1-unit increase) 1.04 (0.92, 1.16)
4 (2 case-control, 2 cross-sectional) NDVI >100-300m buffer 1.00 (0.99, 1.02)
3 (2 cross-sectional, 1 case-control) NDVI >300-500m buffer 1.04 (0.90, 1.22)
Asthma incidence Tang et al. (2023) 9 (4 cross-sectional, 2 case-control, 2 cohort, 1 longitudinal) NDVI RR (per 0.1-unit increase) 0.92 (0.85, 0.98)
Asthma prevalence Tang et al. (2023) 4 (2 cohort, 2 cross-sectional) NDVI RR (per 0.1-unit increase) 0.89 (0.74, 1.08)
COPD j Tang et al. (2023) 4 (3 cross-sectional, 1 cohort) NDVI 500m buffer RR (per 0.1-unit increase) 0.95 (0.89, 1.02)
3 (2 cross-sectional, 1 cohort) NDVI 1000m buffer 0.92 (0.83, 1.03)
COPD incidence Tang et al. (2023) 2 (1 cohort, 1 cross-sectional) NDVI RR (per 0.1-unit increase) 0.92 (0.83, 1.03)
COPD prevalence Tang et al. (2023) 2 (cross-sectional) NDVI RR (per 0.1-unit increase) 1.00 (0.90, 1.12)
Allergic rhinitis Cao et al. (2023) 5 (3 cohort, 2 cross-sectional) NDVI 100m buffer OR (per 10% increase) 1.00 (1.00, 1.00)
2 (1 cohort, 1 cross-sectional) NDVI 200m buffer 1.00 (0.99, 1.01)
3 (2 cohort, 1 cross-sectional) NDVI 250m buffer 1.00 (1.00, 1.00)
3 (cohort) NDVI 300m buffer 1.00 (0.99, 1.02)
5 (3 cohort, 2 cross-sectional) NDVI 500m buffer 0.99 (0.97, 1.01)
2 (1 cohort, 1 cross-sectional) NDVI 1000m buffer 0.99 (0.97, 1.01)
10 (7 cohort, 3 cross-sectional) NDVI 100-1000m buffer 1.00 (0.99, 1.00)
Tang et al. (2023) 5 (2 cohort, 2 cross-sectional, 2 case-control) NDVI 200-300m buffer RR (per 0.1-unit increase) 1.02 (0.95, 1.09)
7 (3 cohort, 3 cross-sectional, 1 cohort) NDVI 500m buffer 0.99 (0.94, 1.03)
4 (2 cross-sectional, 1 cohort) NDVI 1000m buffer 0.99 (0.91, 1.06)
Wang et al. (2022) 2 (cohort) NDVI 100m buffer OR (3 rd tertile vs 1 st tertile) 0.65 (0.27, 1.55)
3 (cohort) NDVI 100-500m buffer 0.75 (0.59, 0.95)
2 (cohort) NDVI 500m buffer 0.91 (0.70, 1.19)
4 (cohort) NDVI 0.83 (0.72, 0.96)
Wu et al. (2022) 3 (2 cross-sectional, 1 case-control) NDVI 100m buffer OR (per 0.1-unit increase) 0.98 (0.95, 1.02)
4 (2 cross-sectional, 1 case-control, 1 cohort) NDVI 500m buffer 0.99 (0.94, 1.04)
3 (2 cross-sectional, 1 case-control) NDVI 1000m buffer 1.00 (0.95, 1.05)
Ye et al. (2022) 6 (3 cohort, 3 cross-sectional) NDVI OR (per 0.1-unit increase) 0.95 (0.85, 1.06)
Allergic rhinitis incidence Tang et al. (2023) 6 (4 cohort, 1 case-control, 1 cross-sectional) NDVI RR (per 0.1-unit increase) 1.02 (0.97, 1.08)
Allergic rhinitis prevalence Tang et al. (2023) 2 (cross-sectional) NDVI RR (per 0.1-unit increase) 0.91 (0.64, 0.95)
Cancer
All-site cancer Li et al. (2023) 6 (cohort) NDVI HR k (per 0.1-unit increase) 0.98 (0.95, 1.01)
Prostate cancer Li et al. (2023) 4 (cohort) NDVI HR (per 0.1-unit increase) 0.95 (0.85, 1.05)
Zare Sakhvidi, Yang, et al. (2022b) 2 (1 cohort, 1 case-control) NDVI OR/RR (per 0.1-unit increase) 0.91 (0.69, 1.20)
2 (cohort) NDVI HR (per 0.1-unit increase) 0.98 (0.69, 1.40)
Colorectal cancer Li et al. (2023) 2 (cohort) NDVI HR (per 0.1-unit increase) 1.00 (0.96, 1.04)
Bladder cancer Li et al. (2023) 2 (cohort) NDVI HR (per 0.1-unit increase) 0.83 (0.45, 1.53)
Breast cancer Li et al. (2023) 5 (cohort) NDVI HR (per 0.1-unit increase) 0.95 (0.90, 1.01)
Zare Sakhvidi, Yang, et al. (2022b) 2 (1 cohort, 1 case-control) NDVI OR/RR (per 0.1-unit increase) 1.01 (0.80, 1.27)
2 (cohort) NDVI HR (per 0.1-unit increase) 0.83 (0.47, 1.48)
Lung cancer Li et al. (2023) 3 (cohort) NDVI HR (per 0.1-unit increase) 0.97 (0.95, 0.98)
Tang et al. (2023) 5 (3 cohort, 1 case-control, 1 ecological) NDVI RR (per 0.1-unit increase) 0.62 (0.40, 0.95)
3 (cohort) NDVI 500m buffer 0.70 (0.46, 1.06)
2 (1 cohort, 1 ecological) NDVI 1000m buffer 0.20 (0.01, 4.48)
Zare Sakhvidi, Yang, et al. (2022b) 4 (cross-sectional) NDVI IRR l (per 0.1-unit increase) 0.99 (0.85, 1.48)
2 (cohort) NDVI HR (per 0.1-unit increase) 1.00 (0.87, 1.38)
2 (1 case-control, 1 cross-sectional) NDVI OR/RR (per 0.1-unit increase) 1.00 (0.84, 1.20)
Skin cancer Li et al. (2023) 2 (cohort) NDVI HR (per 0.1-unit increase) 0.86 (0.57, 1.30)
General health and quality of life
Quality of life Briggs et al. (2022) 7 (RCT) Gardening SMD (intervention vs control) -0.06 (-0.45, 0.34)
Overall wellbeing Briggs et al. (2022) 4 (RCT) Gardening SMD (intervention vs control) 0.37 (0.01, 0.73)
All-cause and cause-specific mortality
All-cause mortality Yuan et al. (2021) 8 (cohort) NDVI HR (per 0.1-unit increase) 0.99 (0.97, 1.00)
Cardiovascular mortality Bianconi et al. (2023) 6 (cohort) NDVI HR (per IQR increase) 0.94 (0.91, 0.97)
Yuan et al. (2021) 4 (cohort) NDVI HR (per 0.1-unit increase) 0.99 (0.89,1.09)
Ischaemic heart disease mortality Bianconi et al. (2023) 5 (cohort) NDVI HR (per IQR increase) 0.96 (0.93, 0.99)
Yuan et al. (2021) 3 (cohort) NDVI HR (per 0.1-unit increase) 0.96 (0.88, 1.05)
Stroke mortality Yuan et al. (2021) 4 (cohort) NDVI HR (per 0.1-unit increase) 0.77 (0.59, 1.00)
Cerebrovascular disease mortality Bianconi et al. (2023) 5 (cohort) NDVI HR (per IQR increase) 0.96 (0.94, 0.97)
Respiratory mortality Yuan et al. (2021) 5 (cohort) NDVI HR (per 0.1-unit increase) 0.99 (0.89, 1.10)
COPD mortality Tang et al. (2023) 3 (2 cohort, 1 ecological) NDVI RR (per 0.1-unit increase) 0.95 (0.92, 0.99)
2 (1 cohort, 1 ecological) NDVI 1000m buffer 0.93 (0.88, 0.98)
Lung cancer mortality Li et al. (2023) 4 (cohort) NDVI HR (per 0.1-unit increase) 0.97 (0.95, 0.98)
Tang et al. (2023) 6 (4 cohort, 2 ecological) NDVI RR (per 0.1-unit increase) 0.98 (0.96, 1.01)
5 (4 cohort, 1 ecological) NDVI 250-300m buffer 0.98 (0.94, 1.01)
4 (cohort) NDVI 1000m buffer 0.98 (0.94, 1.03)
Other health outcomes
Salivary cortisol concentration Qiu et al. (2022) 13 (10 crossover RCT, 3 non-RCT) Forest therapy MD (change in intervention vs control) -0.07 (-0.10, -0.04)

95% CI: 95% confidence interval;

NDVI: Normalised differential vegetation index;

RR: Risk ratio;

OR: Odds ratio;

SMD: Standardised mean difference;

Encompasses residential greenness and decreased visits to greenspace;

Encompasses number of neighbourhood parks, distance to useable green space, distance to total green space proportion of useable green space within 300m, proportion of total green space within 300m, proportion of useable green space within 3km, and proportion of total green space within 3km;

ADHD: Attention deficit hyperactivity disorder;

MD: Mean difference;

COPD: Chronic obstructive pulmonary disorder;

HR: Hazard ratio;

IRR: Incidence rate ratios.

 Quality of evidence and risk of bias

Of the 45 systematic reviews included, most exhibited a high risk of bias (n = 35, 77.8%) ( Figure 1; extended data: Supplementary Table S2) and critically low quality (n = 36, 80.0%) ( Table 6; extended data: Supplementary Figure S1). Only one review was deemed high quality ( Piva et al., 2024). The corrected coverage area (CCA) was calculated as 0.007, indicating minimal overlap in the original studies.

Risk of bias assessment of included reviews using ROBIS.

Rows 1-4 show judgements on individual criteria while the bottom row shows the overall risk of bias.

 Methodological quality assessment of included systematic reviews using AMSTAR-2.
Question number a 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Quality
Author, Year
Aarthi et al. (2023) Y b PY N PY Y N N PY NA/PY N N/NA N Y Y N Y CL
Ahmer et al. (2024) Y Y N N Y Y N PY NA/N N NA/Y Y Y Y Y Y CL
Batterham et al. (2022) Y N N PY Y Y N N NA/N N NA/NA NA Y N NA N CL
Bianconi et al. (2023) Y Y N PY Y N N PY NA/PY N NA/Y Y Y Y Y Y L
Bolanis et al. (2024) Y PY N PY Y Y N PY N/PY N NA/NA NA Y Y NA Y CL
Briggs et al. (2022) Y Y Y PY Y Y N Y Y/NA N Y/NA Y Y Y N Y CL
Browning et al. (2022) Y N N N Y Y N N NA/N N NA/NA NA Y N NA Y CL
Cao et al. (2023) Y N N PY Y Y N PY NA/Y Y NA/Y Y Y Y Y Y CL
Ccami-Bernal et al. (2023) Y PY N PY Y Y N PY NA/PY N NA/NA NA Y Y NA Y L
Chae et al. (2021) Y N N N N Y N PY Y/PY N NA/NA NA N N NA Y CL
Chen et al. (2022) Y N N PY Y N Y PY NA/Y N NA/NA NA Y N NA Y L
De la Fuente et al. (2021) Y N N PY Y Y N N NA/PY N NA/NA NA Y N NA Y CL
Gao et al. (2024) Y PY N PY Y Y N PY Y/PY N NA/NA NA N Y NA Y CL
Hsueh et al. (2022) Y PY N PY N N N N NA/PY N NA/NA NA Y Y NA Y L
Kang et al. (2022) Y N N PY N N N N Y/NA N N/NA N Y Y Y Y CL
Lampert et al. (2021) Y N N PY Y Y N N NA/PY N NA/NA NA N N NA Y CL
Li et al. (2023) Y Y Y PY Y Y Y PY NA/PY Y NA/Y N Y Y N Y L
Li and Lange (2023) Y N N N Y N N N N/N N NA/NA NA N N NA Y CL
Masdor et al. (2023) Y PY N PY Y Y N PY NA/PY N NA/NA NA Y Y NA N L
Patwary et al. (2024) Y Y Y PY Y Y Y PY NA/Y Y NA/Y Y Y Y Y Y H
Piva et al. (2024) Y N N PY Y N N PY N/N N NA/NA NA Y N NA Y CL
Qiu et al. (2022) Y N Y PY Y N N PY Y/PY N Y/N Y Y Y Y Y CL
Rahimi-Ardabili et al. (2021) Y N N PY Y N N PY Y/PY N NA/NA NA N Y NA Y CL
Ricciardi et al. (2022) Y N N PY N N N N NA/PY N NA/NA NA Y N NA Y CL
Rigolon et al. (2021) Y N N PY Y Y N N PY/PY N NA/NA NA Y N NA Y CL
Rojas-Rueda et al. (2021) Y N Y PY Y Y N N NA/PY N NA/NA NA N Y NA Y CL
Siah et al. (2023) Y N N PY Y Y N PY PY/N N Y/N N N Y N Y CL
Sivak et al. (2021) Y N N PY Y N N N N/N N NA/NA NA N N NA N CL
Tang et al. (2023) Y Y N N Y Y N PY NA/PY N NA/Y N N Y Y Y CL
Tharrey and Darmon (2021) Y N N PY N N N PY N/PY N NA/NA NA Y N NA Y CL
Tu (2022) Y N Y PY N Y N PY PY/NA N Y/NA Y Y Y Y Y CL
Wang et al. (2022) Y N N N Y Y N PY NA/PY N NA/Y Y N Y Y Y CL
Wu et al. (2022) Y Y N PY Y Y N PY NA/Y Y NA/Y Y Y Y Y Y L
Ye et al. (2022) Y N Y PY Y Y N PY NA/PY N NA/Y N Y N N N CL
Yi et al. (2022) Y N N PY N N N N Y/NA N Y/NA N N N N Y CL
Yuan et al. (2021) Y Y N N Y Y N PY NA/PY N NA/Y Y N Y Y Y CL
Zagnoli et al. (2022) Y N N PY Y N N PY NA/Y N NA/Y Y N Y N Y CL
Zare Sakhvidi, Knobel, et al. (2022a) Y N N PY Y N Y PY NA/PY N NA/NA NA Y Y NA Y L
Zare Sakhvidi, Yang, et al. (2022b) Y Y N PY Y N Y N NA/PY Y NA/Y N Y Y N Y CL
Zhang et al. (2021) Y N N N N N N PY NA/PY N NA/NA NA N Y NA Y CL
Y. Zhang et al. (2024a) Y N N N N N N PY NA/PY N NA/Y Y Y Y Y Y CL
Y. D. Zhang et al. (2024b) Y N N N Y Y N PY N/Y Y NA/NA NA Y Y NA Y CL
X. Zhang et al. (2022c) Y N N PY Y Y N PY PY/N N Y/N Y N Y Y Y CL
Zhao et al. (2022) Y Y N N Y Y N PY NA/Y Y NA/N Y Y Y Y Y CL
Zhao et al. (2021) Y N N PY Y Y N PY NA/PY N NA/Y Y Y Y Y Y CL

1: Participant, Intervention, Comparison, Outcome (PICO) components, 2: Pre-established protocol, 3: Explanation of included studies’ design, 4: Comprehensive search strategy, 5: Study selection in duplicate, 6: Data extraction in duplicate, 7: List of excluded studies with justification, 8: Description of included studies, 9: Assessment of risk of bias (RoB) in included studies, 10: Funding sources of included studies, 11: Use of appropriate statistical methods, 12: RoB impact on synthesised results, 13: Results interpretation with RoB reference, 14: Heterogeneity explanation, 15: Publication/small study bias investigation, 16: Conflict of interest declaration.

Y = Yes; PY= Partial Yes; N = No; NA = Not applicable (no meta-analysis conducted); CL = Critically low; L = Low; H = High.

 Discussion

This umbrella review builds on prior work ( Bryer et al., 2024), highlighting growing interest in the relationship between greenspace exposure and human health ( Farkas et al., 2023). By summarising findings from 45 systematic reviews published since December 2020, we offer a comprehensive and updated overview of health outcomes linked to greenspace exposure. The results reflect both the breadth of outcomes investigated and the variability in evidence across health domains.

We found beneficial associations between greenspace exposure and a range of health outcomes categories, particularly mental health ( Bonaccorsi et al., 2023; Cuijpers et al., 2023; Zare Sakhvidi et al., 2023), QoL ( Bonaccorsi et al., 2023), and cardiovascular health ( Liu et al., 2023), which were consistent with earlier studies. Compared to our previous umbrella review, which summarised studies published up to December 2020, this review observed stronger associations for cardiovascular and metabolic health, as well as general health and QoL. This review also revealed that favourable effects for birth outcomes, cardiovascular health, and mortality were more pronounced in urban areas than in less urban areas ( Browning et al., 2022). Likewise, associations between greenspace and general and cardiovascular health were more pronounced among individuals with a lower SES than those with a higher SES ( Rigolon et al., 2021). This suggests that greenspace is likely to function as a buffer against urban stressors and contribute to reducing health disparities ( Bressane et al., 2024; Heo & Bell, 2023).

However, we found more inconsistent evidence for all-cause mortality, cognitive function, and cancer outcomes than in our previous review. This shift in the evidence may reflect the expansion of research in some domains. Systematic reviews examining respiratory outcomes such as asthma and AR presented mixed findings - some reporting protective effects, others showing null or conflicting associations. Similar inconsistencies were noted in prior umbrella reviews ( Yang et al., 2021; Zare Sakhvidi et al., 2023). These inconsistencies may be attributable to variability in study designs, greenspace measures, and definitions of health outcomes across original studies. Limited and inconclusive evidence was also observed for maternal health and birth outcomes (beyond BW and PTB) ( Xie et al., 2024) and cancer outcomes ( Yang et al., 2021), restricting the strength of conclusions in these domains.

Since our previous review ( Bryer et al., 2024), we have observed a notable increase in the number of randomised controlled trials (RCTs) evaluating greenspace interventions, particularly in relation to mental health and psychological stress. Commonly studied interventions include forest bathing, community gardening, and nature-based therapy. Although RCTs remain less prevalent in environmental health than observational designs, their emergence represents a positive shift ( Pynegar et al., 2021). Unlike observational designs, which are susceptible to residual confounding and reverse causality, RCTs allow causal inference. Therefore, the findings are more convincing for urban planning and public health interventions ( Gale et al., 2023).

Nonetheless, limitations persist. Many RCTs to date have small sample sizes, short durations, and lack blinding of participants and assessors, which may reduce internal validity ( Pynegar et al., 2021). In addition, the heterogeneity in intervention types, duration, and intensity makes synthesis challenging. Despite these issues, the upward trend in RCTs underscores a growth in the evidence on how greenspace interventions can promote mental health and well-being.

Modest shifts in greenspace exposure measures since our previous umbrella review were also identified. While NDVI remained the most commonly used measure, proximity to greenspace and percentage-based metrics slightly declined since our previous review. Notably, other satellite-derived indices such as the EVI and SAVI were more frequently used. Both of our umbrella reviews highlight the wide, heterogeneous range of greenspace exposure measures used which proves to be an ongoing challenge in cross-study comparability ( Freymueller et al., 2024).

Compared to our previous umbrella review, this review identified a wider breadth of health outcomes studied in relation to greenspace exposure, nearly doubling the number of health outcomes examined. In the previous review, only four systematic reviews focused on cancer outcomes, covering a limited range of cancer types. In contrast, this review identified twice as many studies on cancer, encompassing a broader spectrum of cancer types, including colorectal, breast, and oral cancers. The number of other health outcomes also grew substantially to include communicable diseases such as tuberculosis, dengue, and malaria, reflecting an increasing interest in the relationship between greenspace and infectious disease burden. This notable expansion corroborates the growing volume of research in this field ( Farkas et al., 2023).

Despite these advancements, a decline in the overall quality of systematic reviews has been observed since the onset of the COVID-19 pandemic ( Baumeister et al., 2021). Several systematic reviews included in this umbrella review lacked one or more of the following: preregistered protocols, comprehensive and reproducible search strategies, and adequate reporting of risk of bias and synthesis methods. These shortcomings reflect key departures from established best practices outlined in the ROBIS ( Whiting et al., 2016) and AMSTAR-2 guidelines ( Shea et al., 2017). This raises concerns about the reliability of recent evidence and underscores the need for high-quality systematic reviews.

The strengths of this umbrella review lie in its comprehensive approach. By conducting an extensive search across five databases, it captures a broader and more diverse range of studies compared to prior reviews ( Cuijpers et al., 2023; Xie et al., 2024; Yang et al., 2021; Zare Sakhvidi et al., 2023). The inclusion of both quantitative and qualitative evidence offers a more holistic understanding of the association between greenspace exposure and health outcomes and addresses limitations of earlier reviews that focus on quantitative data ( Xie et al., 2024). Another strength of this umbrella review is its exclusive focus on health outcomes, rather than upstream determinants such as physical activity or social cohesion. This focus ensures clearer attribution of health effects to greenspace exposure itself.

This umbrella review is also subject to several limitations that should be considered when interpreting the findings. Evidence on the association between greenspace and human health is highly heterogeneous due to differences in definitions of greenspace, exposure measures, and outcome measures across original studies and systematic reviews. This diversity limits the comparability of systematic reviews and makes it difficult to draw definitive conclusions. This umbrella review is also impacted by the generally high risk of bias and low methodological quality of included systematic reviews. Finally, while systematic reviews were included or excluded based on publication date, this was not extended to original studies. As such, this umbrella review may include systematic reviews that synthesised original studies published before December 2020, potentially leading to overlap in the original studies covered by this umbrella review and our previous umbrella review.

Mechanisms underlying the health benefits of greenspace exposure remain a critical area of inquiry. Moderators such as urbanicity and SES were explored by Browning et al. (2022) and Rigolon et al. (2021), respectively. However, mediators such as increased social cohesion ( Elliott et al., 2023; L. Zhang et al., 2022b) and decreased environmental stressors ( Chen et al., 2021; Wang & Tassinary, 2024) have been investigated in other studies. For mental health and cognitive function outcomes, interventions like community gardening and forest therapy may enhance social cohesion ( J. Zhang et al., 2022a) and promote relaxation ( Zhang et al., 2021). Social interaction facilitated by greenspaces can foster community bonds and reduce feelings of isolation, further contributing to mental well-being and cognitive function ( Nawrath et al., 2022). The aesthetic qualities of greenspaces, such as visual greenery, may also stimulate positive psychological responses ( Nawrath et al., 2022).

There are a variety of other mechanisms for the association between greenspace and human health outcomes. For cardiovascular and metabolic health, greenspaces may reduce exposure to environmental stressors such as heat and noise, while promoting independent or group physical activities ( Cardinali et al., 2023; Keith et al., 2024; Pan et al., 2024). These activities not only improve physical fitness but also regulate blood pressure, cholesterol levels, and glucose metabolism ( Keith et al., 2024). Additionally, greenspaces may improve air quality by reducing particulate matter and other pollutants, leading to better respiratory ( Li et al., 2024) and cardiovascular ( Qiu et al., 2021) health. Investigation of causal pathways is needed to gain a thorough understanding of these mechanisms.

The combined findings of our umbrella reviews have important implications for urban planning and public health policies. Increasing access to greenspace, particularly in urban and socioeconomically disadvantaged areas, could be a simple and cost-effective strategy to improve population health and reduce health inequalities ( Bressane et al., 2024; Rigolon et al., 2021). Integration of greenspace into healthcare interventions, such as green prescriptions has increased ( Kondo et al., 2020) and should continue to be prioritised alongside other initiatives including urban greening projects and preservation of existing greenspaces ( Bikomeye et al., 2021). Ensuring community consultation, equitable distribution, and long-term maintenance of green spaces will also be essential to maximising health benefits. This will require collaboration across sectors, including health, environment, transport, and housing, to create green spaces that are both resilient and responsive to local needs.

Author contributions

Brittnee Bryer: methodology, validation, formal analysis, investigation, data curation, writing – original draft, writing – review & editing, visualisation. Nicholas Osborne: conceptualisation, validation, methodology, writing – review & editing, supervision. Jialu Wang: investigation, data curation, validation, writing – review & editing. Rajarshi Dasgupta: validation, writing – review & editing. Gail Williams: conceptualisation, methodology, writing – review & editing, supervision. Darsy Darssan: conceptualisation, methodology, validation, formal analysis, investigation, data curation, writing – original draft, writing – review & editing, visualisation, supervision, project administration.

 Data availability Underlying data

All data underlying the results are available as part of the article and no additional source data are required.

 Extended data

Open Science Framework (OSF): Greenspace exposure and associated health outcomes: an updated systematic review of reviews. Supplementary Material. DOI: http://doi.org/10.17605/OSF.IO/XPZCJ

This project contains the following extended data:

Supplementary Table S1. (Literature search strategies)

Supplementary Table S2. (Risk of bias assessment by domain)

Supplementary Figure S1: (Overall methodological quality)

Supplementary File S1. (Systematic reviews excluded during full-text screening)

Reporting guidelines

Open Science Framework (OSF): PRISMA checklist and flow chart for Greenspace exposure and associated health outcomes: an updated systematic review of reviews. DOI: 10.17605/OSF.IO/XPZCJ

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

 Software availability

Endnote version 20 was used as a reference management tool in this study ( The EndNote Team, 2013). A free alternative to Endnote is Zotero ( Zotero, 2024).

Covidence was used to streamline eligibility screening and data extraction in this study ( Covidence systematic review software, 2022). A free alternative to Covidence in Rayyan ( Ouzzani et al., 2016).

 References Aarthi GR Mehreen Begum TS Moosawi SA : Associations of the built environment with type 2 diabetes in Asia: a systematic review. BMJ Open. 2023;13(4):e065431. 37015791 10.1136/bmjopen-2022-065431 PMC10083821 Ahmer Z Atif M Zaheer S : Association between residential green spaces and pregnancy outcomes: a systematic review and meta-analysis. Int. J. Environ. Health Res. 2024;34:3188–3205. 38185100 10.1080/09603123.2023.2299242 Aromataris E Fernandez R Godfrey CM : Summarizing systematic reviews: methodological development, conduct and reporting of an umbrella review approach. JBI Evid. Implement. 2015;13(3). 38185100 10.1097/XEB.0000000000000055 Batterham PJ Brown K Trias A : Systematic review of quantitative studies assessing the relationship between environment and mental health in rural areas. Aust. J. Rural Health. 2022;30(3):306–320. 35189016 10.1111/ajr.12851 PMC9303895 Baumeister A Corrin T Abid H : The quality of systematic reviews and other synthesis in the time of COVID-19. Epidemiol. Infect. 2021;149:e182. 34340726 10.1017/s0950268821001758 PMC8365192 Bianconi A Longo G Coa AA : Impacts of Urban Green on Cardiovascular and Cerebrovascular Diseases-A Systematic Review and Meta-Analysis. Int. J. Environ. Res. Public Health. 2023;20(11). 37297570 10.3390/ijerph20115966 PMC10253108 Bikomeye JC Namin S Anyanwu C : Resilience and Equity in a Time of Crises: Investing in Public Urban Greenspace Is Now More Essential Than Ever in the US and Beyond. Int. J. Environ. Res. Public Health. 2021;18(16). 34444169 10.3390/ijerph18168420 PMC8392137 Bolanis D Vergunst F Mavoa S : Association between greenspace exposure and suicide-related outcomes across the lifespan: A systematic review. Sci. Total Environ. 2024;906:167451. 37777126 10.1016/j.scitotenv.2023.167451 Bonaccorsi G Milani C Giorgetti D : Impact of Built Environment and Neighborhood on Promoting Mental Health, Well-being, and Social Participation in Older People: an Umbrella Review. Ann. Ig. 2023;35(2):213–239. 35788248 10.7416/ai.2022.2534 Bressane A Cunha Pinto JPd Castro Medeiros LCd : Urban green space disparities: Implications of environmental injustice for public health. Urban For. Urban Green. 2024;99:128441. 10.1016/j.ufug.2024.128441 Briggs R Morris PG Rees K : The effectiveness of group-based gardening interventions for improving wellbeing and reducing symptoms of mental ill-health in adults: a systematic review and meta-analysis. J. Ment. Health. 2022;32:787–804. 36151719 10.1080/09638237.2022.2118687 Browning MHEM Rigolon A McAnirlin O : Where greenspace matters most: A systematic review of urbanicity, greenspace, and physical health. Landsc. Urban Plan. 2022;217:104233. 10.1016/j.landurbplan.2021.104233 Bryer B Odebeatu CO Lee WR : Greenspace exposure and associated health outcomes: a systematic review of reviews [version 1; peer review: 2 approved with reservations]. F1000Res. 2024;13(491). 10.12688/f1000research.148878.1 Bryer B Wang J Williams G : An updated umbrella review: green space exposure and human health outcomes. PROSPERO 2022 CRD42022383421. 2022. Retrieved 04/09/2023. Reference Source Cao NW Zhou HY Du YJ : The effect of greenness on allergic rhinitis outcomes in children and adolescents: A systematic review and meta-analysis. Sci. Total Environ. 2023;859(Pt 1):160244. 36402344 10.1016/j.scitotenv.2022.160244 Cardinali M Balderrama A Arztmann D : Green walls and health: An umbrella review. Nature-Based Solutions. 2023;3:100070. 10.1016/j.nbsj.2023.100070 Ccami-Bernal F Soriano-Moreno DR Fernandez-Guzman D : Green space exposure and type 2 diabetes mellitus incidence: A systematic review. Health Place. 2023;82:103045. 37159977 10.1016/j.healthplace.2023.103045 Chae Y Lee S Jo Y : The Effects of Forest Therapy on Immune Function. Int. J. Environ. Res. Public Health. 2021;18(16). 34444188 10.3390/ijerph18168440 PMC8394293 Chen K Zhang T Liu F : How Does Urban Green Space Impact Residents’ Mental Health: A Literature Review of Mediators. Int. J. Environ. Res. Public Health. 2021;18(22). 34831512 10.3390/ijerph182211746 PMC8621109 Chen P-W Chen L-K Huang H-K : Productive Aging by Environmental Volunteerism: A Systematic Review. Arch. Gerontol. Geriatr. 2022;98:104563. 34741895 10.1016/j.archger.2021.104563 Covidence systematic review software: Veritas Health Innovation. 2022. Reference Source Cuijpers P Miguel C Ciharova M : Impact of climate events, pollution, and green spaces on mental health: an umbrella review of meta-analyses. Psychol. Med. 2023;53(3):638–653. 36606450 10.1017/S0033291722003890 PMC9975983 De la Fuente F Saldías MA Cubillos C : Green space exposure association with type 2 diabetes mellitus, physical activity, and obesity: A systematic review. Int. J. Environ. Res. Public Health. 2021;18(1):1–18. 33375559 10.3390/ijerph18010097 PMC7796153 Elliott LR Pasanen T White MP : Nature contact and general health: Testing multiple serial mediation pathways with data from adults in 18 countries. Environ. Int. 2023;178:108077. 37413929 10.1016/j.envint.2023.108077 Farkas JZ Hoyk E Morais MB de : A systematic review of urban green space research over the last 30 years: A bibliometric analysis. Heliyon. 2023;9(2):e13406. 36816272 10.1016/j.heliyon.2023.e13406 PMC9932659 Freymueller J Schmid HL Senkler B : Current methodologies of greenspace exposure and mental health research-a scoping review. Front. Public Health. 2024;12:1360134. 38510363 10.3389/fpubh.2024.1360134 PMC10951718 Gale RP Zhang M-J Lazarus HM : The role of randomized controlled trials, registries, observational databases in evaluating new interventions. Best Pract. Res. Clin. Haematol. 2023;36(4):101523. 38092482 10.1016/j.beha.2023.101523 Gao Y Pan K Li H : Greenspace Exposure with Chronic Obstructive Pulmonary Disease: A Systematic Review. Forests. 2024;15(4). 10.3390/f15040634 Heo S Bell ML : Investigation on urban greenspace in relation to sociodemographic factors and health inequity based on different greenspace metrics in 3 US urban communities. J. Expo. Sci. Environ. Epidemiol. 2023;33(2):218–228. 35995844 10.1038/s41370-022-00468-z PMC10005950 Hsueh YC Batchelor R Liebmann M : A Systematic Review of Studies Describing the Effectiveness, Acceptability, and Potential Harms of Place-Based Interventions to Address Loneliness and Mental Health Problems. Int. J. Environ. Res. Public Health. 2022;19(8). 35457637 10.3390/ijerph19084766 PMC9029472 Kang MJ Kim HS Kim JY : Effects of Forest-Based Interventions on Mental Health: A Meta-Analysis of Randomized Controlled Trials. Int. J. Environ. Res. Public Health. 2022;19(8). 35457755 10.3390/ijerph19084884 PMC9031621 Keith RJ Hart JL Bhatnagar A : Greenspaces And Cardiovascular Health. Circ. Res. 2024;134(9):1179–1196. 38662868 10.1161/CIRCRESAHA.124.323583 Khalaf RKS Akaraci S Baldwin FD : Causally inferred evidence of the impact of green and blue spaces (GBS) on maternal and neonatal health: a systematic review and meta-analysis. Environmental Research: Health. 2025;3. 10.1088/2752-5309/ada8ce Kondo MC Oyekanmi KO Gibson A : Nature Prescriptions for Health: A Review of Evidence and Research Opportunities. Int. J. Environ. Res. Public Health. 2020;17(12). 32545651 10.3390/ijerph17124213 PMC7344564 Lampert T Costa J Santos O : Evidence on the contribution of community gardens to promote physical and mental health and well-being of non-institutionalized individuals: A systematic review. PLoS ONE. 2021;16(8):e0255621. 34358279 10.1371/journal.pone.0255621 PMC8345884 Li J Xie Y Xu J : Association between greenspace and cancer: evidence from a systematic review and meta-analysis of multiple large cohort studies. Environ. Sci. Pollut. Res. 2023;30(39):91140–91157. 37474858 10.1007/s11356-023-28461-5 Li K Mao Y Li Y : Exploring the pathways of urban green space exposure on respiratory health: An empirical study in Nanjing, China. Urban For. Urban Green. 2024;101:128536. 10.1016/j.ufug.2024.128536 Li L Lange KW : Assessing the Relationship between Urban Blue-Green Infrastructure and Stress Resilience in Real Settings: A Systematic Review. Sustainability. 2023;15(12). 10.3390/su15129240 Liu M Meijer P Lam TM : The built environment and cardiovascular disease: an umbrella review and meta-meta-analysis. Eur. J. Prev. Cardiol. 2023;30(16):1801–1827. 37486178 10.1093/eurjpc/zwad241 Masdor NA Abu Bakar MF Hod R : Green space exposure and colorectal cancer: A systematic review. Heliyon. 2023;9(5):e15572. 37153430 10.1016/j.heliyon.2023.e15572 PMC10160744 Moher D Liberati A Tetzlaff J : Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. The BMJ. 2009;339:b2535. 19622551 10.1136/bmj.b2535 PMC2714657 Nawrath M Elsey H Dallimer M : Why cultural ecosystem services matter most: Exploring the pathways linking greenspaces and mental health in a low-income country. Sci. Total Environ. 2022;806:150551. 34627115 10.1016/j.scitotenv.2021.150551 Ouzzani M Hammady H Fedorowicz Z : Rayyan — a web and mobile app for systematic reviews. 2016;5(210). 10.1186/s13643-016-0384-4 Pan R Wang W Wei N : Does the morphology of residential greenspaces contribute to the development of a cardiovascular-healthy city? Environ. Res. 2024;257:119280. 38821460 10.1016/j.envres.2024.119280 Patwary MM Sakhvidi MJZ Ashraf S : Impact of green space and built environment on metabolic syndrome: A systematic review with meta-analysis. Sci. Total Environ. 2024;923:170977. 38360326 10.1016/j.scitotenv.2024.170977 Piva G Caruso L Gómez AC : Effects of forest walking on physical and mental health in elderly populations: a systematic review. Rev. Environ. Health. 2024;39(1):121–136. 36239186 10.1515/reveh-2022-0093 Pollock M Fernandes RM Becker LA : Overviews of Reviews.In: Higgins JPT, Thomas J, Chandler J, et al.(editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.5. Cochrane,2024[last updated August 2023]. Reference Source Pynegar EL Gibbons JM Asquith NM : What role should randomized control trials play in providing the evidence base for conservation? Oryx. 2021;55(2):235–244. 10.1017/S0030605319000188 Qiu Q Yang L He M : The Effects of Forest Therapy on the Blood Pressure and Salivary Cortisol Levels of Urban Residents: A Meta-Analysis. Int. J. Environ. Res. Public Health. 2022;20(1). 36612777 10.3390/ijerph20010458 PMC9819785 Qiu Y Zuo S Yu Z : Discovering the effects of integrated green space air regulation on human health: A bibliometric and meta-analysis. Ecol. Indic. 2021;132:108292. 10.1016/j.ecolind.2021.108292 Rahimi-Ardabili H Astell-Burt T Nguyen PY : Green Space and Health in Mainland China: A Systematic Review. Int. J. Environ. Res. Public Health. 2021;18(18). 34574854 10.3390/ijerph18189937 PMC8472560 Ricciardi E Spano G Lopez A : Long-Term Exposure to Greenspace and Cognitive Function during the Lifespan: A Systematic Review. Int. J. Environ. Res. Public Health. 2022;19(18). 36141977 10.3390/ijerph191811700 PMC9517665 Rigolon A Browning M McAnirlin O : Green Space and Health Equity: A Systematic Review on the Potential of Green Space to Reduce Health Disparities. Int. J. Environ. Res. Public Health. 2021;18(5). 33806546 10.3390/ijerph18052563 PMC7967323 Rojas-Rueda D Vaught E Buss D : Why a new research agenda on green spaces and health is needed in Latin America: Results of a systematic review. Int. J. Environ. Res. Public Health. 2021;18(11). 34072319 10.3390/ijerph18115839 PMC8198896 Sharifi Y Sobhani S Ramezanghorbani N : Association of greenspaces exposure with cardiometabolic risk factors: a systematic review and meta-analysis. BMC Cardiovasc. Disord. 2024;24(1):170. 38509487 10.1186/s12872-024-03830-1 PMC10953288 Shea BJ Reeves BC Wells G : AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both. The BMJ. 2017;358:j4008. 28935701 10.1136/bmj.j4008 PMC5833365 Siah CJR Goh YS Lee J : The effects of forest bathing on psychological well-being: A systematic review and meta-analysis. Int. J. Ment. Health Nurs. 2023;32(4):1038–1054. 36864583 10.1111/inm.13131 Sivak CJ Pearson AL Hurlburt P : Effects of vacant lots on human health: A systematic review of the evidence. Landsc. Urban Plan. 2021;208:104020. 10.1016/j.landurbplan.2020.104020 Song J Gasparrini A Wei D : Do greenspaces really reduce heat health impacts? Evidence for different vegetation types and distance-based greenspace exposure. Environ. Int. 2024;191:108950. 39190977 10.1016/j.envint.2024.108950 Tang M Liu W Li H : Greenness and chronic respiratory health issues: a systematic review and meta-analysis. Front. Public Health. 2023;11:1279322. 38125839 10.3389/fpubh.2023.1279322 PMC10732026 Tharrey M Darmon N : Urban collective garden participation and health: a systematic literature review of potential benefits for free-living adults. Nutr. Rev. 2021;80(1):6–21. 33997887 10.1093/nutrit/nuaa147 The EndNote Team: EndNote. (Version EndNote 20) [64-bit]. Clarivate;2013. Tu HM : Effect of horticultural therapy on mental health: A meta-analysis of randomized controlled trials. J. Psychiatr. Ment. Health Nurs. 2022;29(4):603–615. 35000249 10.1111/jpm.12818 Wang H Gholami S Xu W : Where and how to invest in greenspace for optimal health benefits: a systematic review of greenspace morphology and human health relationships. The Lancet Planetary Health. 2024;8(8):e574–e587. 39122326 10.1016/S2542-5196(24)00140-2 Wang H Tassinary LG : Association between greenspace morphology and prevalence of non-communicable diseases mediated by air pollution and physical activity. Landsc. Urban Plan. 2024;242:104934. 10.1016/j.landurbplan.2023.104934 Wang X Zhou N Zhi Y : Association between exposure to greenness and atopic march in children and adults-A systematic review and meta-analysis. Front. Public Health. 2022;10:1097486. 36699899 10.3389/fpubh.2022.1097486 PMC9868616 Whiting P Savovic J Higgins JP : ROBIS: A new tool to assess risk of bias in systematic reviews was developed. J. Clin. Epidemiol. 2016;69:225–234. 26092286 10.1016/j.jclinepi.2015.06.005 PMC4687950 World Health Organization: International statistical classification of diseases and related health problems. 2019. Retrieved 04/09/2023. Reference Source Wu B Guo X Liang M : Association of individual green space exposure with the incidence of asthma and allergic rhinitis: a systematic review and meta-analysis. Environ. Sci. Pollut. Res. 2022;29(59):88461–88487. 36329245 10.1007/s11356-022-23718-x Xie Y Fan S Luo Y : Credibility of the evidence on green space and human health: an overview of meta-analyses using evidence grading approaches. EBioMedicine. 2024;106:105261. 39079340 10.1016/j.ebiom.2024.105261 PMC11340586 Yang B-Y Zhao T Hu L-X : Greenspace and human health: An umbrella review. The Innovation. 2021;2(4):100164. 34622241 10.1016/j.xinn.2021.100164 PMC8479545 Ye T Yu P Wen B : Greenspace and health outcomes in children and adolescents: A systematic review. Environ. Pollut. 2022;314:120193. 36122655 10.1016/j.envpol.2022.120193 Yi Y Seo E An J : Does Forest Therapy Have Physio-Psychological Benefits? A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Int. J. Environ. Res. Public Health. 2022;19(17). 36078228 10.3390/ijerph191710512 PMC9518146 Yuan Y Huang F Lin F : Green space exposure on mortality and cardiovascular outcomes in older adults: a systematic review and meta-analysis of observational studies. Aging Clin. Exp. Res. 2021;33(7):1783–1797. 32951189 10.1007/s40520-020-01710-0 Zagnoli F Filippini T Jimenez MP : Is Greenness Associated with Dementia? A Systematic Review and Dose-Response Meta-analysis. Curr. Environ. Health Rep. 2022;9:574–590. 35857243 10.1007/s40572-022-00365-5 PMC9729322 Zare Sakhvidi MJ Knobel P Bauwelinck M : Greenspace exposure and children behavior: A systematic review. Sci. Total Environ. 2022a;824:153608. 35134416 10.1016/j.scitotenv.2022.153608 Zare Sakhvidi MJ Mehrparvar AH Zare Sakhvidi F : Greenspace and health, wellbeing, physical activity, and development in children and adolescents: An overview of the systematic reviews. Curr. Opin. Environ. Sci. Health. 2023;32:100445. 10.1016/j.coesh.2023.100445 Zare Sakhvidi MJ Yang J Mehrparvar AH : Exposure to greenspace and cancer incidence, prevalence, and mortality: A systematic review and meta-analyses. Sci. Total Environ. 2022b;838:156180. 35618130 10.1016/j.scitotenv.2022.156180 Zhang J Liu Y Zhou S : Do various dimensions of exposure metrics affect biopsychosocial pathways linking green spaces to mental health? A cross-sectional study in Nanjing, China. Landsc. Urban Plan. 2022a;226:104494. 10.1016/j.landurbplan.2022.104494 Zhang L Tan PY Gan DRY : Assessment of mediators in the associations between urban green spaces and self-reported health. Landsc. Urban Plan. 2022b;226:104503. 10.1016/j.landurbplan.2022.104503 Zhang R Zhang CQ Rhodes RE : The pathways linking objectively-measured greenspace exposure and mental health: A systematic review of observational studies. Environ. Res. 2021;198:111233. 33933490 10.1016/j.envres.2021.111233 Zhang X Zhang Y Yun J : A systematic review of the anxiety-alleviation benefits of exposure to the natural environment. Rev. Environ. Health. 2022c;38:281–293. 35334194 10.1515/reveh-2021-0157 Zhang Y Wu T Yu H : Green spaces exposure and the risk of common psychiatric disorders: A meta-analysis. SSM - Population Health. 2024a;25:101630. 38405164 10.1016/j.ssmph.2024.101630 PMC10885792 Zhang YD Zhou GL Wang L : Greenspace and human microbiota: A systematic review. Environ. Int. 2024b;187:108662. 38653130 10.1016/j.envint.2024.108662 Zhao Y Bao WW Yang BY : Association between greenspace and blood pressure: A systematic review and meta-analysis. Sci. Total Environ. 2022;817:152513. 35016929 10.1016/j.scitotenv.2021.152513 Zhao YL Qu Y Ou YN : Environmental factors and risks of cognitive impairment and dementia: A systematic review and meta-analysis. Ageing Res. Rev. 2021;72:101504. 34755643 10.1016/j.arr.2021.101504 Zotero: (Version 6.0.36) Corporation for Digital Scholarship. 2024. Reference Source 10.5256/f1000research.198281.r485819 Reviewer response for version 2 Senkler Ben 1 Referee https://orcid.org/0000-0001-8518-4782 Freymueller Julius 1 Co-referee Universitat Bielefeld Medizinische Fakultat OWL (Ringgold ID: 575404), Bielefeld, North Rhine Westfalia, Germany

Competing interests: No competing interests were disclosed.

 15 6 2026 Copyright: © 2026 Senkler B and Freymueller J 2026 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

Thank you for giving us the opportunity to conduct a review of the manuscript ““ Greenspace exposure and associated health outcomes: an updated systematic review of reviews”. We reviewed this submission for the first time at version 2 and did not assess version 1, so the comments below represent an initial overall appraisal of the current version rather than a follow-up on earlier remarks.

This is a substantial and carefully executed umbrella review, highlighting important aspects of the interaction of greenspace exposure and health. The authors searched five databases, synthesised both quantitative and qualitative systematic reviews, and covered a broad range of health outcomes and age groups, which makes the manuscript a useful and up-to-date map of the greenspace–health literature. The dual appraisal with AMSTAR-2 and ROBIS, the ICD-10 mapping of outcomes, and the candid reporting of the predominantly low methodological quality of the evidence base are strengths of this umbrella review. In Version 2, the authors have also clearly responded earlier reviewer remarks, the expanded Tables 1 and 5, the per-outcome summary sentences, the CCA calculation, and the extended limitations are all well considered additions. We have several remaining minor comments, which we believe could further strengthen the manuscript.

Corrected coverage area (CCA): the overall value is hard to interpret. We agree with the point raised independently by Dr Anna Soerensen and Dr Mingwei Liu. A single overall CCA of 0.007 says little here, and can even be misleading. With 45 reviews spread across eight outcome categories (most covering only one or two), a low overall overlap is to be expected and does not really show that the reviews are independent. What matters is the overlap within a category, where the same primary studies can recur and make the evidence look more consistent than it is. We would therefore recommend reporting the CCA at least per outcome category (Per individual outcome would be cleaner in principle but unstable with so few reviews each, so the category level is the practical compromise.) The overall value can stay as a supplement, but with a note that a low figure is expected given the category structure.

The single high-quality review appears to be misidentified. The Results text ("Quality of evidence and risk of bias") states that only one review was rated high quality and names Piva et al. (2024). But Table 6 shows the single AMSTAR-2 "High" rating for Patwary et al. (2024) and rates Piva et al. (2024) as "Critically low". The text and Table 6 are therefore inconsistent. Please check the consistency of the text with Table 6.

Inconsistent characterisation of maternal and birth outcomes across sections. The conclusion for this category differs across the three places it is mentioned/summarized: the Abstract lists it under “ inconsistent associations”. The summary sentence at the end of the Results subsection states that the evidence showed protective associations but the Discussion describes it as limited and inconclusivebeyond birthweight and preterm birth. We would ask the authors to harmonise the wording across the Abstract, Results, and Discussion to ensure consistence.

The predominantly positive framing does not clearly reflect the documented potential harms.The "Other health outcomes" subsection and Table 4 report several adverse or risk-increasing associations (for example elevated blood lead levels in children and a higher prevalence of West Nile virus-carrying mosquitoes around green vacant lots, and an increased incidence of malaria, dengue, and tuberculosis in one review, alongside dermatophytosis and toxocariasis). These are not reflected in the Abstract or in the overall framing of the Discussion, which present greenspace as uniformly beneficial. Inconsistent results are transparently addressed in the discussion, but we would encourage the authors to consider acknowledging the documented potential harms, so that the manuscript presents an even more balanced picture.

 Small adaptations

Table 3 denominator. The column header reads "% out of 44 reviews," but the percentages are calculated on 45 reviews (e.g., NDVI: 27/45 = 60.0%; 27/44 would be 61.4%). The header should be corrected to 45.

Causal/evaluative language. "Protective," and "favourable" are used throughout the Abstract, Results, and Discussion to describe associations drawn from an evidence base that is largely cross-sectional (45.7% of original studies: Table 2) for example, and several subsections describe "protective" effects. "Protective" in particular implies a causal, risk-reducing relationship that cross-sectional data cannot establish. We would invite the authors to consider softening some of this wording toward more neutral associational language, and to decide for themselves how far to adjust it.

Cortisol terminology and the CI. In "Other health outcomes," the salivary cortisol result from Qiu et al. (2022) is described as "serum cortisol concentration (SCC)" with a 95% CI of (-1.10, -0.04), whereas Table 5 labels the same result "salivary cortisol concentration" with a CI of (-0.10, -0.04); the abbreviation list also defines SCC as serum cortisol. As far as we see, the underlying study measured salivary cortisol, so "serum" appears to be an error, and the lower CI bound in the text (-1.10) is inconsistent with Table 5 (-0.10). Please check.

Cognitive function. The Abstract groups cognition within "mental health" (rated beneficial), whereas the Discussion lists "cognitive function" among the outcomes for which the evidence was more inconsistent. The signaling on cognition could be made consistent between the two sections.

Maternal/birth review count text versus Table 4. The Results state that five reviews investigated maternal health and birth outcomes and name Rojas-Rueda et al. (2021) among them, but the maternal/birth section of Table 4 lists only four reviews (Rojas-Rueda's outcome, infant mortality, appears under the mortality category). Please check if you would like to adjust it.

Distinction between intervention and ambient-exposure evidence. The summary verdicts for some categories (mental health, for example) draw on two quite different kinds of evidence: observational studies of residential or surrounding greenspace (e.g., NDVI in cross-sectional studies) and intervention studies such as forest bathing or gardening (often randomised). These differ both in the type of exposure and in how strongly they support causal claims, so a single combined verdict can blur that distinction. The authors may want to consider whether and how to separate these two strands or mention this conflation.

Reporting guideline. The Methods cite PRISMA 2009 (Moher et al., 2009). For an overview of reviews, the dedicated PRIOR statement (Gates et al., 2022) and/or the updated PRISMA 2020 statement (Page et al., 2021) are the current reporting standards, and adherence to one of these would be preferable.

Citation of ROBIS. In the Methods, ROBIS is cited as "(Shea et al., 2017)"; ROBIS is Whiting et al. (2016) (as correctly cited later in the Discussion and in the reference list) while Shea et al. (2017) is the AMSTAR-2 reference. Please adjust the citation accordingly.

Addition of physical activity as an important mediating factor. The Discussion presents a comprehensive overview of moderators and mediators regarding the impact of greenspace exposure on health outcomes. Physical activity is mentioned here only as a relevant mechanism for cardiovascular and metabolic health, although it is stated in the Results as a mediator for mental health. Please consider adding physical activity to the discussion of potential mediators.

Are the rationale for, and objectives of, the Systematic Review clearly stated?

Yes

Is the statistical analysis and its interpretation appropriate?

Yes

If this is a Living Systematic Review, is the ‘living’ method appropriate and is the search schedule clearly defined and justified? (‘Living Systematic Review’ or a variation of this term should be included in the title.)

Not applicable

Are sufficient details of the methods and analysis provided to allow replication by others?

Yes

Are the conclusions drawn adequately supported by the results presented in the review?

Partly

Reviewer Expertise:

Environmental (mental) Health / environmental epidemiology

We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however we have significant reservations, as outlined above.

 10.5256/f1000research.198281.r485825 Reviewer response for version 2 Hudgins Emma J 1 Referee https://orcid.org/0000-0002-8402-5111 The University of Melbourne, Parkville, Australia

Competing interests: No competing interests were disclosed.

 15 6 2026 Copyright: © 2026 Hudgins EJ 2026 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

This article by Bryer and colleagues conducts a review of reviews on the evidence for associations between a range of health outcomes and exposure to green spaces. This umbrella review is an update to an existing umbrella review and follows the same methods. It includes qualitative findings as well as a summary of all meta-analytic results available on these associations. The authors found that greenspace exposure had well supported positive links to mental health, general health and quality of life, and cardiovascular and metabolic health. Greenspace had more variable associations with cancer, respiratory health and allergies, and maternal health and birth outcomes. In spite of these strong linkages described across included reviews, the authors caution that methodological rigour is low across included studies and is in decline.

This article is strong in its methodological reproducibility and adherence to FAIR principles- especially given the pre-registration of its protocol and following the revisions requested by the first round of reviewers e.g. including the search string in this article rather than only pointing a previous article. I also recognize the review’s novelty in combining synthesis from both conventional medical studies and more interdisciplinary analyses.  Its message around continuing to strive for preregistration and other important research practices that have declined in recent years is an important one and the authors here practice what they preach. The slight weakness of the paper are the need for a bit more methodological information and wording that seems to suggest a hierarchy of disciplinary knowledge surrounding the issue being studied.

More specifically, beyond the additional concerns raised by the other reviewers, I have two of my own  minor concerns

1) the V1 reviewer 1 comment regarding whether you conducted forward and backward review of references contained within reviews was not addressed by the revised wording around search terms in the supplementary table. Could you add additional clarification around this?

2) The discussion section sets up RCTs as the ideal study design, potentially unfairly, given the challenges in holistically capturing green space exposure in a feasible RCT. I think lines of evidence spanning a range of study designs adds a degree of holism to the umbrella review and could be framed more positively.

Are the rationale for, and objectives of, the Systematic Review clearly stated?

Yes

Is the statistical analysis and its interpretation appropriate?

Partly

If this is a Living Systematic Review, is the ‘living’ method appropriate and is the search schedule clearly defined and justified? (‘Living Systematic Review’ or a variation of this term should be included in the title.)

Not applicable

Are sufficient details of the methods and analysis provided to allow replication by others?

Yes

Are the conclusions drawn adequately supported by the results presented in the review?

Yes

Reviewer Expertise:

Urban ecology, biodiversity conservation

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

 10.5256/f1000research.198281.r476283 Reviewer response for version 2 Soerensen Anna 1 Referee https://orcid.org/0000-0001-6393-0940 Hasselt University, Hasselt, Belgium

Competing interests: No competing interests were disclosed.

 8 5 2026 Copyright: © 2026 Soerensen A 2026 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

I thank the authors for implementing my previous suggestions. The revisions have enhanced the clarity of the manuscript, making it straightforward to follow throughout. However, I still have one comment: Could the authors clarify the rationale for calculating the overall CCA rather than per category? Low overall overlap is somewhat expected, as most review studies only span one or two categories out of the 8 included. This might not accurately reflect the degree of overlap within each category, where it's most relevant for interpreting the evidence.

Are the rationale for, and objectives of, the Systematic Review clearly stated?

Yes

Is the statistical analysis and its interpretation appropriate?

Yes

If this is a Living Systematic Review, is the ‘living’ method appropriate and is the search schedule clearly defined and justified? (‘Living Systematic Review’ or a variation of this term should be included in the title.)

Not applicable

Are sufficient details of the methods and analysis provided to allow replication by others?

Partly

Are the conclusions drawn adequately supported by the results presented in the review?

Partly

Reviewer Expertise:

Environmental epidemiology

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

 10.5256/f1000research.198281.r476281 Reviewer response for version 2 Liu Mingwei 1 Referee https://orcid.org/0000-0003-3984-1280 Huazhong University of Science and Technology, Wuhan, China

Competing interests: No competing interests were disclosed.

 23 4 2026 Copyright: © 2026 Liu M 2026 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

I appreciate that the authors have nicely addressed my previous comments. I still have one comment regarding the quality and bias assessment. 

This is feedback on points 3 and 5 of my previous comments.  

To point 3: The CCA should be calculated for each outcome. Because the results were synthesized for each outcome. The CCA shall demonstrate the degree of overlap of primary studies that was synthesized. And if there is a large overlap for an outcome, authors can choose the input studies. 

To point 5: The results were interpreted for each health outcome. Therefore, authors need to know the quality and bias of each result by outcome. The quality and bias should be analyzed for each outcome, so that it is more informative for future studies.

Are the rationale for, and objectives of, the Systematic Review clearly stated?

Partly

Is the statistical analysis and its interpretation appropriate?

Yes

If this is a Living Systematic Review, is the ‘living’ method appropriate and is the search schedule clearly defined and justified? (‘Living Systematic Review’ or a variation of this term should be included in the title.)

Not applicable

Are sufficient details of the methods and analysis provided to allow replication by others?

Partly

Are the conclusions drawn adequately supported by the results presented in the review?

Yes

Reviewer Expertise:

Environmental epidemiology

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

 10.5256/f1000research.183900.r444035 Reviewer response for version 1 Soerensen Anna 1 Referee https://orcid.org/0000-0001-6393-0940 Hasselt University, Hasselt, Belgium

Competing interests: No competing interests were disclosed.

 13 1 2026 Copyright: © 2026 Soerensen A 2026 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

Thank you for giving me the opportunity to review the manuscript entitled “Greenspace exposure and associated health outcomes: an updated systematic review of reviews”. This umbrella review provides an overview of the latest reviews within the field of green space exposure and health outcomes. Overall, the manuscript succeeds in providing this overview and clearly reflects the substantial effort invested in its preparation. It also highlights the important observation that the vast majority of included systematic reviews and meta-analyses are of low quality. Additionally, the structure and writing are clear and well organized, making the manuscript a pleasure to read. However, I do have some comments.

Major comments:

The search strategy is not entirely clear. Referring to your previous review article gives the impression that the full search strategy can be found there. For instance, in your previous review you state: “To ensure that relevant systematic reviews were retrieved, we also conducted forward and backward reference searching on screened reviews, as well as manual searching of reference lists of relevant reviews”, which makes it unclear whether this was also conducted for the current manuscript. I would also recommend explicitly stating that the search terms and strategies can be found in Supplementary Table S1, and perhaps providing a few examples of the search terms used.

This is a well-conducted umbrella review, but it would be markedly strengthened by an assessment of study overlap. If no assessment of study overlap is conducted, I would suggest including a statement in the limitations section and/or providing a rationale for why this was not considered relevant.

Including age groups and geographical location in Table 1 would further strengthen the manuscript, particularly as coverage across all age groups is one of the rationales for conducting this review.

 Minor comments:

The abstract is well written, to the point, and correctly summarizes the findings of the manuscript. My only suggestion is to add  “green space” and " greenness" as keywords, if possible.

In the methods section, I would appreciate clarification regarding “iv) reported health outcome(s) directly attributable to greenspace exposure”. 

The results section has a logical flow and is easy to follow. However, I am missing a definition of “overall greenspace exposure” and “forest-based interventions”. It is not clear which types of studies are included under these terms.

On page 4, the sentence “with mediating factors such as air quality, perceived stress, and physical activity identified (Zhang et al., 2021)” is included, but Zhang et al. (2021) is not mentioned in the preceding sentence, making it unclear which type of green space exposure they assessed.

I would consider adding one or two sentences at the end of each outcome subsection summarizing the overall findings. Almost all outcome subsections include at least one review article that does not align with the findings of the other reviews, which makes it difficult to determine whether there is an overall effect. While you do summarize the main findings and conclusions in the discussion, I believe it would be helpful to include such summary statements throughout the results section, for example at the end of the “Cardiovascular and metabolic outcomes” subsection.

The findings suggesting that associations may be moderated by SES, urbanicity, or related contextual factors are interesting and add nuance to the results. However, as this aspect was not described in the methods section, its inclusion in the results came somewhat unexpectedly. Clarifying this in the methods and/or presenting these findings in a dedicated subsection of the results would improve  readability

On page 34, you state: “Similarly, a 0.1-unit increment in NDVI was associated with a reduced risk of asthma incidence (Tang et al., 2023)”, which gives the impression that an odds ratio or another effect estimate is missing.

On page 35, “Gardening interventions were associated with better physical health” was stated. Could you clarify how physical health was defined?

Are the rationale for, and objectives of, the Systematic Review clearly stated?

Yes

Is the statistical analysis and its interpretation appropriate?

Yes

If this is a Living Systematic Review, is the ‘living’ method appropriate and is the search schedule clearly defined and justified? (‘Living Systematic Review’ or a variation of this term should be included in the title.)

Not applicable

Are sufficient details of the methods and analysis provided to allow replication by others?

Partly

Are the conclusions drawn adequately supported by the results presented in the review?

Partly

Reviewer Expertise:

Environmental epidemiology

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Darssan Darsy The University of Queensland, Australia

Competing interests: No competing interests.

 31 3 2026

We are thankful for your thorough and valuable comments on our manuscript. We have carefully reviewed and integrated your recommendations to the extent possible.

Thank you for giving me the opportunity to review the manuscript entitled “Greenspace exposure and associated health outcomes: an updated systematic review of reviews”. This umbrella review provides an overview of the latest reviews within the field of green space exposure and health outcomes. Overall, the manuscript succeeds in providing this overview and clearly reflects the substantial effort invested in its preparation. It also highlights the important observation that the vast majority of included systematic reviews and meta-analyses are of low quality. Additionally, the structure and writing are clear and well organized, making the manuscript a pleasure to read. However, I do have some comments.

Major comments:

1. The search strategy is not entirely clear. Referring to your previous review article gives the impression that the full search strategy can be found there. For instance, in your previous review you state: “To ensure that relevant systematic reviews were retrieved, we also conducted forward and backward reference searching on screened reviews, as well as manual searching of reference lists of relevant reviews”, which makes it unclear whether this was also conducted for the current manuscript. I would also recommend explicitly stating that the search terms and strategies can be found in Supplementary Table S1, and perhaps providing a few examples of the search terms used.

We have updated the methods section to ensure that the location of the search strategy is stated more explicitly. 

2. This is a well-conducted umbrella review, but it would be markedly strengthened by an assessment of study overlap. If no assessment of study overlap is conducted, I would suggest including a statement in the limitations section and/or providing a rationale for why this was not considered relevant.

We have calculated the CCA value and have added this and its interpretation to the article.

3. Including age groups and geographical location in Table 1 would further strengthen the manuscript, particularly as coverage across all age groups is one of the rationales for conducting this review.

We have added the age range studies in each systematic review to the fifth column of Table 1. We have also added the location(s) covered by each systematic review to the second column of Table 1. Due to the large number and wide variety of countries covered by systematic reviews, we chose to report study location by continent instead.

 

Minor comments:

1. The abstract is well written, to the point, and correctly summarizes the findings of the manuscript. My only suggestion is to add “green space” and " greenness" as keywords, if possible.

Thank you for this suggestion.

2. In the methods section, I would appreciate clarification regarding “iv) reported health outcome(s) directly attributable to greenspace exposure”.

We have clarified this criterion in the methods section by noting that eligible reviews needed to examine health outcomes specifically attributable to greenspace exposure, independent of blue space.

3. The results section has a logical flow and is easy to follow. However, I am missing a definition of “overall greenspace exposure” and “forest-based interventions”. It is not clear which types of studies are included under these terms.

We have added an explanation of what these terms encompass.

4. On page 4, the sentence “with mediating factors such as air quality, perceived stress, and physical activity identified (Zhang et al., 2021)” is included, but Zhang et al. (2021) is not mentioned in the preceding sentence, making it unclear which type of green space exposure they assessed.

We have revised this section to improve clarity.

5. I would consider adding one or two sentences at the end of each outcome subsection summarizing the overall findings. Almost all outcome subsections include at least one review article that does not align with the findings of the other reviews, which makes it difficult to determine whether there is an overall effect. While you do summarize the main findings and conclusions in the discussion, I believe it would be helpful to include such summary statements throughout the results section, for example at the end of the “Cardiovascular and metabolic outcomes” subsection.

We have added a sentence at the end of each health outcome section to summarise the results.

6. The findings suggesting that associations may be moderated by SES, urbanicity, or related contextual factors are interesting and add nuance to the results. However, as this aspect was not described in the methods section, its inclusion in the results came somewhat unexpectedly. Clarifying this in the methods and/or presenting these findings in a dedicated subsection of the results would improve readability

The findings related to SES and urbanicity as moderators of the association between greenspace and human health outcomes were not part of the predefined objectives of the umbrella review. For this reason, we have not added them to the methods section or expanded their presentation in the results section. We have kept them only as brief contextual notes to avoid over‑interpreting the evidence.

7. On page 34, you state: “Similarly, a 0.1-unit increment in NDVI was associated with a reduced risk of asthma incidence (Tang et al., 2023)”, which gives the impression that an odds ratio or another effect estimate is missing.

We have added the effect estimate for this association.

8. On page 35, “Gardening interventions were associated with better physical health” was stated. Could you clarify how physical health was defined?

Physical health was generally self-reported and encompassed perceived general health, acute health complaints, physical constraints, and chronic illnesses.

 10.5256/f1000research.183900.r433260 Reviewer response for version 1 Liu Mingwei 1 Referee https://orcid.org/0000-0003-3984-1280 Huazhong University of Science and Technology, Wuhan, China

Competing interests: No competing interests were disclosed.

 27 12 2025 Copyright: © 2025 Liu M 2025 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

The current umbrella review updated the current evidence of the association between greenspace exposure and various health outcomes. The review is conducted in a systematic approach and has a comprehensive coverage of evidence. However, some measures like corrected covered area (CCA), robustness of evidence, and time scale can be added when synthesizing the evidence. The implications should be more specific. 

Major

1. Are there reviews of a topic (health outcome) that were identified only in the previous umbrella review you conducted, but not in this updated review? 

2. This umbrella review builds upon your previous review. It is good to specify that the previous review included publications between January 2010 and December 2020 at the beginning. 

3. Have you considered the overlap of primary studies in the included systematic review? It is usually assessed by a measure of CCA. If some reviews had a high level of overlap of primary studies, you may consider only including one of them. 

4. The study should provide the specific search strategy (the combination of search terms) in a supplementary file or clearly refer to where to find it. 

5. The methodological quality, risk of bias, and robustness of each evidence are not reported in the Results text. It is clearer to label them alongside the evidence. It is the same for table 1 and table 5. Table 6 can be put into the supplementary file. The overall quality, risk of bias, and robustness of each evidence should be made clear in the corresponding text and tables. 

6. What about the robustness of each quantitative evidence? The robustness comprehensively reflects the magnitude of the p-value, the sample size, the heterogeneity, and the risk of bias. There is a simple rule for assessing robustness suggested before. Please refer to the literature below. 

    1. Aromataris E, Fernandez R, Godfrey CM, Holly C, Khalil H, Tungpunkom P. Summarizing systematic reviews: methodological development, conduct and reporting of an umbrella review approach. Int J Evid Based Healthc 2015;13:132–140.

    2. Aromataris E, Fernandez RS, Godfrey CHolly C, Khalil H, Tungpunkom P. Methodology for JBI umbrella reviews 2014;1.

    3. Cleophas TJ, Zwinderman AH. Meta-meta-analysis. In: Modern meta-analysis: review and update of methodologies. Cham: Springer International Publishing; 2017, p135–143.

    4. Fusar-Poli P, Radua J. Ten simple rules for conducting umbrella reviews. Evid Based Ment Health 2018;21:95–100.

7. Even though the included systematic reviews are different between the current umbrella review and the previous umbrella review, the included original studies (evidence) of the current umbrella review should cover most of the original studies in the previous umbrella review. Because the included systematic review did not set the criteria to include only original studies between December 2020 and June 2024, rather, they should have included earlier original studies. How do you interpret this? The readers should be aware of it when interpreting results. 

8. The current review included a wide range of diseases and various study designs. The effective exposure time of green space may vary among diseases. For example, the time scale of the association between green space and CVD is usually years, whereas the time scale of the association between green space and mental health can be days. This should be reported in tables and results. And it could be an interesting discussion point. 

9. What is the limitation of the current study? 

10. The authors should give more implications for policy and future studies regarding health domains, respectively. 

Minor

1. In the first paragraph of the Introduction, you said "However, limitations in the methodology of these umbrella reviews have resulted in some gaps in the literature." But not all of the listed reviews are umbrella review. You can consider to rephrase it to "these umbrella or systematic reviews". 

2. In the second paragraph of the Introduction, you first said "these reviews often present their findings narratively", then you gave an example review that "focused solely on quantitative results". It is not clear what you suggest from here? Should study include both quantitative and qualitative analyses? Or maybe you want to demonstrate that previous reviews did not include both quantitative and qualitative results. But you should give more example reviews. 

3. Some texts are not well cited. For example, in the Discussion section, "Systematic reviews examining respiratory outcomes such as asthma and AR presented mixed findings - some reporting protective effects, others showing null or conflicting associations". Please examine thoroughly to make sure the texts have citations where it is necessary.

Are the rationale for, and objectives of, the Systematic Review clearly stated?

Partly

Is the statistical analysis and its interpretation appropriate?

Yes

If this is a Living Systematic Review, is the ‘living’ method appropriate and is the search schedule clearly defined and justified? (‘Living Systematic Review’ or a variation of this term should be included in the title.)

Not applicable

Are sufficient details of the methods and analysis provided to allow replication by others?

Partly

Are the conclusions drawn adequately supported by the results presented in the review?

Yes

Reviewer Expertise:

Environmental epidemiology

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Darssan Darsy The University of Queensland, Australia

Competing interests: No competing interests.

 31 3 2026

We sincerely appreciate the thorough and insightful feedback on the manuscript. We have carefully considered and incorporated your suggestions wherever feasible.

The current umbrella review updated the current evidence of the association between greenspace exposure and various health outcomes. The review is conducted in a systematic approach and has a comprehensive coverage of evidence. However, some measures like corrected covered area (CCA), robustness of evidence, and time scale can be added when synthesizing the evidence. The implications should be more specific.

Major

1. Are there reviews of a topic (health outcome) that were identified only in the previous umbrella review you conducted, but not in this updated review?

While the major health outcome categories remained consistent with those in our previous umbrella review, some individual outcomes did not appear in this umbrella review. These include gestational age, pregnancy complications, head circumference, atherosclerosis, hypotension, oxidative stress, and inflammation markers. However, it is important to note that these outcomes may have been investigated under larger categories, such as overall cardiovascular health and physiological outcomes, by systematic reviews.

2. This umbrella review builds upon your previous review. It is good to specify that the previous review included publications between January 2010 and December 2020 at the beginning.

We have added the date range of our previous umbrella review to the methods section.

3. Have you considered the overlap of primary studies in the included systematic review? It is usually assessed by a measure of CCA. If some reviews had a high level of overlap of primary studies, you may consider only including one of them.

We have calculated the CCA value and have added this and its interpretation to the results section under Quality of Evidence and Risk of Bias.

4. The study should provide the specific search strategy (the combination of search terms) in a supplementary file or clearly refer to where to find it.

The full search strategy is available in the extended data section (per the F1000 guidelines) as Supplementary Table 1. We have added a sentence in the methods section to make finding the full search strategy clearer. 

5. The methodological quality, risk of bias, and robustness of each evidence are not reported in the Results text. It is clearer to label them alongside the evidence. It is the same for table 1 and table 5. Table 6 can be put into the supplementary file. The overall quality, risk of bias, and robustness of each evidence should be made clear in the corresponding text and tables.

Because quality and risk‑of‑bias ratings apply to the systematic reviews as a whole (not to each specific outcome), linking them directly to individual findings could be misleading. For this reason, we present these assessments separately and discuss their implications in the narrative synthesis.

6. What about the robustness of each quantitative evidence? The robustness comprehensively reflects the magnitude of the p-value, the sample size, the heterogeneity, and the risk of bias. There is a simple rule for assessing robustness suggested before. Please refer to the literature below.

Aromataris E, Fernandez R, Godfrey CM, Holly C, Khalil H, Tungpunkom P. Summarizing systematic reviews: methodological development, conduct and reporting of an umbrella review approach. Int J Evid Based Healthc 2015;13:132–140.

Aromataris E, Fernandez RS, Godfrey CHolly C, Khalil H, Tungpunkom P. Methodology for JBI umbrella reviews 2014;1.

Cleophas TJ, Zwinderman AH. Meta-meta-analysis. In: Modern meta-analysis: review and update of methodologies. Cham: Springer International Publishing; 2017, p135–143.

Fusar-Poli P, Radua J. Ten simple rules for conducting umbrella reviews. Evid Based Ment Health 2018;21:95–100.

We assessed the methodological quality and risk of bias of all included systematic reviews using both AMSTAR-2 and ROBIS. These validated tools evaluate key domains relevant to the robustness of evidence, including risk of bias, heterogeneity, and overall methodological rigour.

The references above primarily recommend critical appraisal of the included systematic reviews and transparent reporting of heterogeneity, potential biases, and study characteristics. None of these sources explicitly mandates an additional composite "robustness" index that integrates p-value magnitude, sample size, heterogeneity, and risk of bias for each quantitative result beyond standard quality appraisal tools.

7. Even though the included systematic reviews are different between the current umbrella review and the previous umbrella review, the included original studies (evidence) of the current umbrella review should cover most of the original studies in the previous umbrella review. Because the included systematic review did not set the criteria to include only original studies between December 2020 and June 2024, rather, they should have included earlier original studies. How do you interpret this? The readers should be aware of it when interpreting results.

We have added a discussion of this to the limitations section of the discussion.

8. The current review included a wide range of diseases and various study designs. The effective exposure time of green space may vary among diseases. For example, the time scale of the association between green space and CVD is usually years, whereas the time scale of the association between green space and mental health can be days. This should be reported in tables and results. And it could be an interesting discussion point.

We agree that exposure timeframes may differ across diseases. However, many of the systematic reviews included in our umbrella review do not report exposure duration and deriving or assigning exposure timeframes would require re‑interpreting primary studies, which is beyond the scope of an umbrella review. For these reasons, we have not added exposure‑time information to the tables or results.

9. What is the limitation of the current study?

We have added a paragraph to more explicitly discuss the limitations of this umbrella review.

10. The authors should give more implications for policy and future studies regarding health domains, respectively.

While we agree that policy and research implications are important, it is difficult to provide meaningful domain‑specific implications without content analysis of existing policy worldwide.

Minor

1. In the first paragraph of the Introduction, you said "However, limitations in the methodology of these umbrella reviews have resulted in some gaps in the literature." But not all of the listed reviews are umbrella review. You can consider to rephrase it to "these umbrella or systematic reviews".

We have adjusted the wording in this section for clarity.

2. In the second paragraph of the Introduction, you first said "these reviews often present their findings narratively", then you gave an example review that "focused solely on quantitative results". It is not clear what you suggest from here? Should study include both quantitative and qualitative analyses? Or maybe you want to demonstrate that previous reviews did not include both quantitative and qualitative results. But you should give more example reviews.

This paragraph has been revised for clarity.

3. Some texts are not well cited. For example, in the Discussion section, "Systematic reviews examining respiratory outcomes such as asthma and AR presented mixed findings – some reporting protective effects, others showing null or conflicting associations". Please examine thoroughly to make sure the texts have citations where it is necessary.

The statement in question summarises findings that are already fully referenced in the Results section. Because the Discussion section synthesises previously cited material rather than introducing new evidence, we have not repeated citations here. However, we have re‑checked the manuscript to ensure that all statements requiring citation are appropriately referenced.

 10.5256/f1000research.183900.r435773 Reviewer response for version 1 Rojas-Rueda David 1 Referee https://orcid.org/0000-0001-5854-2484 Colorado State University, Fort Collins, USA

Competing interests: No competing interests were disclosed.

 24 12 2025 Copyright: © 2025 Rojas-Rueda D 2025 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

Feedback for Improvement:

1. Transparency of Study Design in Meta-Analyses: The results derived from meta-analyses pooling data primarily from cross-sectional or ecological studies are highly problematic for inferring causality or defining risk, as they are "only useful to define a hypothesis, not to identify that relationship". Even though the paper provides a narrative discussion of the overall study design distribution (Table 2), Table 5 should be enhanced to specify the underlying study designs for each quantitative finding.

    ◦ For example, Bianconi et al. (2023) summarized cardiovascular outcomes pulling from 15 ecological, 13 cohort, 7 cross-sectional, and 1 case-control study. If their quantitative estimates in Table 5 are derived from this mixed pool, a risk assessor cannot easily distinguish the quality of the pooled result.

    ◦ Recommendation: Clarify in the text or in an expanded version of Table 5 the type and proportion of original studies (e.g., Cohort/RCT vs. Cross-sectional/Ecological) contributing to each meta-analysis result. This prevents the perpetuation of bias where the meta-analysis result might be interpreted as a strong causal effect when the underlying evidence base is correlational.

B. Enhancing Dose-Response Information (Table 5)

Table 5, listing quantitative data from meta-analyses, is the most crucial section for PRA. However, it currently lacks the standardization needed for clear risk modeling.

Dose-Response Functions and Exposure Units:

1. Dose Definition and Units: Many entries in Table 5 rely on the Normalised Difference Vegetation Index (NDVI). The protective effect should be explicitly linked to a defined unit of change in exposure (the dose).

    ◦ The table labels the Exposure Measure generally (e.g., "NDVI 500m buffer") but rarely defines the exposure increment associated with the effect estimate (e.g., OR/HR per 0.1 unit increase in NDVI).

    ◦ One example where relative dose information is present is Ahmer et al. (2024), where the standardized regression coefficient (beta) is listed for Birth weight (BW) relative to NDVI 250m and 500m buffers. Similarly, Zhao et al. (2022) provides standardized regression coefficients for systolic and diastolic BP in relation to "Overall greenspace" exposure.

    ◦ Recommendation: Systematically list the implied "dose" unit for all non-binary exposures (i.e., NDVI, Area of greenspace, etc.) in Table 5. If the original meta-analysis reported relative risk for quartile comparison, this should be noted (e.g., "Highest Quartile NDVI vs. Lowest Quartile"). If the study provided results per standard deviation or a standard incremental increase (like the 0.1-unit increment in NDVI mentioned in the narrative for asthma incidence), this information must be made explicit within the table.

2. Missing Contextual Information (Population and Geography)

For effective public health targeting, understanding which populations and regions derive the most benefit is essential.

A. Population Specificity (Age Groups and Health Status)

While the review states that it aims to encompass all age groups and mentions studies focused on children and adolescents and older adults, this information is not consistently integrated into the core summary tables.

1. Age and Health Status Context in Tables: The utility of a finding for a risk assessor depends heavily on the population studied. For instance, the protective effect of greenspace exposure on mortality might be different in a cohort of healthy adults versus older adults with pre-existing cardiovascular conditions.

    ◦ Yuan et al. (2021) focused on mortality and cardiovascular outcomes in older adults.

    ◦ Ye et al. (2022) focused on health outcomes in childhood and adolescence.

    ◦ Recommendation: Systematically add a column to Table 1 (or Table 5 for quantitative data) explicitly identifying the primary population group (e.g., Children/Adolescents, Adults, Older Adults, Mixed) studied in the included systematic reviews. This improves contextual relevance for targeted policy recommendations.

B. Geographical Context and Disparities

The paper notes crucial findings regarding socioeconomic status (SES) and urbanicity, such as stronger protective associations in urban areas compared to less urban areas, and larger benefits for individuals with lower SES. This is highly important for addressing health equity (a key component of modern environmental health).

1. Geographical Reporting: The review includes Rojas-Rueda et al. (2021) which specifically addresses evidence in Latin America. Similarly, Rahimi-Ardabili et al. (2021) covers green space and health in Mainland China. However, the primary geographical focus of other contributing reviews is not consistently highlighted in the data summary tables.

    ◦ Recommendation: Include a column in Table 1 that explicitly identifies the geographic region or primary focus population (e.g., China, Latin America, Global/Mixed HIC) for each included systematic review. This information is vital, allowing policymakers to evaluate generalizability.

3. Missing Key Literature

Rojas-Rueda D, Nieuwenhuijsen MJ, Gascon M, Perez-Leon D, Mudu P. Green spaces and mortality: a systematic review and meta-analysis of cohort studies. (REFER TO 1)

Finding: This review, published in 2019, appears to be missing from the current umbrella review, which restricted its search to articles published between December 2020 and June 2024.

Critique on Exclusion: While adherence to the established search protocol (December 2020 onward) is understandable for an update to a previous review, the excluded Rojas-Rueda (2019) paper is a systematic review and meta-analysis of cohort studies specifically on mortality, an outcome discussed in detail here. Because cohort studies provide stronger evidence for causality than cross-sectional studies (which dominate the current base), and because the authors are looking to update knowledge based on previous work (Bryer et al., 2024), it is essential to contextualize the current findings against high-quality pre-2020 meta-analyses that remain highly relevant.

Recommendation: The authors should clarify if the 2019 Rojas-Rueda mortality review was included in their prior umbrella review (Bryer et al., 2024) and, if so, explicitly reference it in the narrative (e.g., when discussing all-cause mortality findings based on cohort data) to maintain a complete historical context of the highest-quality evidence available in this domain. If this 2019 meta-analysis (focusing solely on cohort data) contradicted or strongly reinforced the results summarized from Yuan et al. (2021) and Bianconi et al. (2023), noting this distinction would significantly improve the paper's utility.

Summary of Key Recommendations for Improving Utility

To maximize the utility of this excellent work for public health policy, I strongly recommend the following improvements:

1. Enhance Table 5 (Quantitative Data): Systematically include columns detailing:

    ◦ The unit or increment of exposure corresponding to the effect estimate (Dose-Response function).

    ◦ The primary study designs (e.g., % Cohort/RCT vs. % Cross-sectional/Ecological) underlying the meta-analysis result.

2. Enhance Table 1 (Review Characteristics): Systematically include columns detailing:

    ◦ The primary demographic group studied (e.g., Children/Adolescents, Older Adults, Mixed Population).

    ◦ The geographic region or scope (e.g., China, Latin America, Global/HIC).

3. Narrative Contextualization: Ensure that the discussion of high-quality pre-2020 systematic reviews (like the 2019 Rojas-Rueda mortality paper) is provided, justifying how the current "update" relates to established, high-causality evidence.

These additions would transform the paper into a far more actionable resource for environmental epidemiologists and urban planners focused on targeted green space policies and health risk quantification.

Are the rationale for, and objectives of, the Systematic Review clearly stated?

Yes

Is the statistical analysis and its interpretation appropriate?

Partly

If this is a Living Systematic Review, is the ‘living’ method appropriate and is the search schedule clearly defined and justified? (‘Living Systematic Review’ or a variation of this term should be included in the title.)

Partly

Are sufficient details of the methods and analysis provided to allow replication by others?

Yes

Are the conclusions drawn adequately supported by the results presented in the review?

Partly

Reviewer Expertise:

Environmental epidemiology

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

References : Green spaces and mortality: a systematic review and meta-analysis of cohort studies. The Lancet Planetary Health .2019;3(11) : 10.1016/S2542-5196(19)30215-3 e469-e477 10.1016/S2542-5196(19)30215-3 Darssan Darsy The University of Queensland, Australia

Competing interests: No competing interests.

 31 3 2026

We are grateful for the valuable and detailed comments on our manuscript. We have thoughtfully reviewed and integrated your recommendations to the extent possible.

Transparency of Study Design in Meta-Analyses: The results derived from meta-analyses pooling data primarily from cross-sectional or ecological studies are highly problematic for inferring causality or defining risk, as they are "only useful to define a hypothesis, not to identify that relationship". Even though the paper provides a narrative discussion of the overall study design distribution (Table 2), Table 5 should be enhanced to specify the underlying study designs for each quantitative finding.

For example, Bianconi et al. (2023) summarized cardiovascular outcomes pulling from 15 ecological, 13 cohort, 7 cross-sectional, and 1 case-control study. If their quantitative estimates in Table 5 are derived from this mixed pool, a risk assessor cannot easily distinguish the quality of the pooled result.

Recommendation: Clarify in the text or in an expanded version of Table 5 the type and proportion of original studies (e.g., Cohort/RCT vs. Cross-sectional/Ecological) contributing to each meta-analysis result. This prevents the perpetuation of bias where the meta-analysis result might be interpreted as a strong causal effect when the underlying evidence base is correlational.

We have added the study designs and numbers to the third column of Table 5.

Enhancing Dose-Response Information (Table 5)

Table 5, listing quantitative data from meta-analyses, is the most crucial section for PRA. However, it currently lacks the standardization needed for clear risk modeling.

Dose-Response Functions and Exposure Units:

Dose Definition and Units: Many entries in Table 5 rely on the Normalised Difference Vegetation Index (NDVI). The protective effect should be explicitly linked to a defined unit of change in exposure (the dose).

The table labels the Exposure Measure generally (e.g., "NDVI 500m buffer") but rarely defines the exposure increment associated with the effect estimate (e.g., OR/HR per 0.1 unit increase in NDVI).

One example where relative dose information is present is Ahmer et al. (2024), where the standardized regression coefficient (beta) is listed for Birth weight (BW) relative to NDVI 250m and 500m buffers. Similarly, Zhao et al. (2022) provides standardized regression coefficients for systolic and diastolic BP in relation to "Overall greenspace" exposure.

Recommendation: Systematically list the implied "dose" unit for all non-binary exposures (i.e., NDVI, Area of greenspace, etc.) in Table 5. If the original meta-analysis reported relative risk for quartile comparison, this should be noted (e.g., "Highest Quartile NDVI vs. Lowest Quartile"). If the study provided results per standard deviation or a standard incremental increase (like the 0.1-unit increment in NDVI mentioned in the narrative for asthma incidence), this information must be made explicit within the table.

We have added the dose or categories being compared to the fifth column of Table 5.

Missing Contextual Information (Population and Geography)

For effective public health targeting, understanding which populations and regions derive the most benefit is essential.

Population Specificity (Age Groups and Health Status)

While the review states that it aims to encompass all age groups and mentions studies focused on children and adolescents and older adults, this information is not consistently integrated into the core summary tables.

Age and Health Status Context in Tables: The utility of a finding for a risk assessor depends heavily on the population studied. For instance, the protective effect of greenspace exposure on mortality might be different in a cohort of healthy adults versus older adults with pre-existing cardiovascular conditions.

Yuan et al. (2021) focused on mortality and cardiovascular outcomes in older adults.

Ye et al. (2022) focused on health outcomes in childhood and adolescence.

Recommendation: Systematically add a column to Table 1 (or Table 5 for quantitative data) explicitly identifying the primary population group (e.g., Children/Adolescents, Adults, Older Adults, Mixed) studied in the included systematic reviews. This improves contextual relevance for targeted policy recommendations.

We have added the age range of studies in each systematic review to the fifth column of Table 1.

Geographical Context and Disparities

The paper notes crucial findings regarding socioeconomic status (SES) and urbanicity, such as stronger protective associations in urban areas compared to less urban areas, and larger benefits for individuals with lower SES. This is highly important for addressing health equity (a key component of modern environmental health).

Geographical Reporting: The review includes Rojas-Rueda et al. (2021) which specifically addresses evidence in Latin America. Similarly, Rahimi-Ardabili et al. (2021) covers green space and health in Mainland China. However, the primary geographical focus of other contributing reviews is not consistently highlighted in the data summary tables.

Recommendation: Include a column in Table 1 that explicitly identifies the geographic region or primary focus population (e.g., China, Latin America, Global/Mixed HIC) for each included systematic review. This information is vital, allowing policymakers to evaluate generalizability.

We have added the location(s) covered by each systematic review to the second column of Table 1. Due to the large number and wide variety of countries covered by systematic reviews, we chose to report study location by continent instead.

Missing Key Literature

Rojas-Rueda D, Nieuwenhuijsen MJ, Gascon M, Perez-Leon D, Mudu P. Green spaces and mortality: a systematic review and meta-analysis of cohort studies. (REFER TO 1)

Finding: This review, published in 2019, appears to be missing from the current umbrella review, which restricted its search to articles published between December 2020 and June 2024.

Critique on Exclusion: While adherence to the established search protocol (December 2020 onward) is understandable for an update to a previous review, the excluded Rojas-Rueda (2019) paper is a systematic review and meta-analysis of cohort studies specifically on mortality, an outcome discussed in detail here. Because cohort studies provide stronger evidence for causality than cross-sectional studies (which dominate the current base), and because the authors are looking to update knowledge based on previous work (Bryer et al., 2024), it is essential to contextualize the current findings against high-quality pre-2020 meta-analyses that remain highly relevant.

Recommendation: The authors should clarify if the 2019 Rojas-Rueda mortality review was included in their prior umbrella review (Bryer et al., 2024) and, if so, explicitly reference it in the narrative (e.g., when discussing all-cause mortality findings based on cohort data) to maintain a complete historical context of the highest-quality evidence available in this domain. If this 2019 meta-analysis (focusing solely on cohort data) contradicted or strongly reinforced the results summarized from Yuan et al. (2021) and Bianconi et al. (2023), noting this distinction would significantly improve the paper's utility.

The scope of this umbrella review was restricted to systematic reviews published between December 2020 and June 2024. As a result, the systematic review by Rojas-Rueda et al. (2019) fell outside the predefined eligibility window and was therefore not included. The study by Rojas-Rueda et al. (2019) was, however, included in our previous umbrella review, which considered systematic reviews published between January 2010 and December 2020. Because this is not a living review and the time window was established a priori, we did not re-include studies published before December 2020 in this umbrella review.

Summary of Key Recommendations for Improving Utility

To maximize the utility of this excellent work for public health policy, I strongly recommend the following improvements:

Enhance Table 5 (Quantitative Data): Systematically include columns detailing:

The unit or increment of exposure corresponding to the effect estimate (Dose-Response function).

The primary study designs (e.g., % Cohort/RCT vs. % Cross-sectional/Ecological) underlying the meta-analysis result.

Enhance Table 1 (Review Characteristics): Systematically include columns detailing:

The primary demographic group studied (e.g., Children/Adolescents, Older Adults, Mixed Population).

The geographic region or scope (e.g., China, Latin America, Global/HIC).

Narrative Contextualization: Ensure that the discussion of high-quality pre-2020 systematic reviews (like the 2019 Rojas-Rueda mortality paper) is provided, justifying how the current "update" relates to established, high-causality evidence.

These additions would transform the paper into a far more actionable resource for environmental epidemiologists and urban planners focused on targeted green space policies and health risk quantification.

References

1. Rojas-Rueda D, Nieuwenhuijsen M, Gascon M, Perez-Leon D, et al.: Green spaces and mortality: a systematic review and meta-analysis of cohort studies. The Lancet Planetary Health. 2019; 3 (11): e469-e477

Thank you for summarising your recommendations. We have responded to each comment and recommendation separately above.