Does the biodiversity of urban green spaces impact on human health? A systematic review of literature Authors: Nicola Pelizzari, Carlotta Alias, Michela Tiboni, Anna Bertolazzi, Claudia Zani Published: 2026 ABSTRACT Green spaces and their biodiversity seem to improve the urban life quality. However, a wide range of descriptors were associated with multiple health outcomes, which limits the comparability of existing evidence. The objective of this systematic review was to organise this knowledge, investigating the relationship between biodiversity in urban green spaces and health outcomes in adults. Following PRISMA guidelines and a registered protocol (PROSPERO: CRD42025636281), a search of PubMed, Scopus, Web of Science identified studies published between January 2014 and July 2025. The search strategy (“biodiversity” AND “human health” AND “urban green space”) yielded 147 eligible studies. Most were published between 2017 and 2020 (57%) and predominantly relied on medium-sized cohorts of 101–10,000 participants (55%). Health outcomes fell into four categories: (i) general health, mental health, and well-being; (ii) drug prescriptions; (iii) chronic diseases (cardiovascular disease, T2 diabetes, obesity); and (iv) birth outcomes. Green exposure was primarily assessed through vegetation-based indicators: 66.1% of studies used green-exposure metrics (tree canopy or vegetation cover), 34.0% NDVI, and 22.4% percentage green area. Only a minority incorporated composite ecological indices or fauna-based metrics (bird or insect richness). Across all health domains, findings were predominantly favourable. Between 70% and 90% of studies on general health, mental health, and well-being reported beneficial associations with green exposure, including improved self-rated health, reduced depressive and anxiety symptoms, lower stress, and increased life satisfaction. All studies using drug prescription data found inverse associations between green exposure and antidepressant or anxiolytic prescriptions. Evidence for chronic diseases consistently indicated protective associations: reduced cardiovascular and respiratory disease risk, prevalence of T2 diabetes, and all-cause mortality. Birth outcomes demonstrated more variability, with half of the studies reporting positive associations, largely influenced by socioeconomic and contextual factors. Socioeconomic status frequently modified associations, with stronger positive effects observed among lower-SES populations. The overall evidence indicates that richness of urban green spaces is associated with health benefits. However, the role of biodiversity remains unclear and understudied, mostly due to inconsistent measurements and ambiguous associations with health outcomes. Integrating ecological quality into urban planning may therefore represent an important strategy for promoting healthier and more equitable urban environments.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12889-026-28039-z. Introduction Current projections indicate that by the year 2050, approximately 70% of the global population will be residing in urban areas [1]. There is broad consensus that the presence and creation of green spaces has the potential to enhance the quality of life for citizens living in urban settings [2]. Urban characteristics, such as the extent and accessibility of green areas, and the quantity and quality of tree cover have been demonstrated to be associated with population health [3, 4]. Furthermore, there is a consistent evidence base that suggests that spending time in urban green spaces has a positive effect on well-being [5, 6]. However, several aspects of health promotion remain insufficiently addressed within urban planning frameworks, such as the enhancement of green spaces in terms of biodiversity [7–9]. This constitutes a health-promoting characteristic of natural environments and a key asset for urban public health [10–12]. Indeed, the lack of biodiversity in green areas is a factor that may increase the risk of disease transmission and other adverse health outcomes [2, 13]. Several mechanisms may explain the health benefits of green areas, including psychological restoration and stress reduction, increased levels of physical activity among populations living near parks, reduced exposure to air pollution, noise, and heat, and enhanced resilience to pollution and climate change [8, 14]. Numerous studies show that higher biodiversity in urban green spaces, whether plant or animal, has favourable effects on mental health, inflammatory processes, several chronic degenerative diseases, and birth outcomes [15–17]. Health effects of green spaces can be assessed using several methodological approaches. Surveys and interviews enable subjective evaluations, whereas biological sampling (e.g., salivary cortisol) allows the measurement of physiological stress responses that may persist for hours after awakening in individuals exposed to green environments [18, 19]. Health-service indicators such as antidepressant prescriptions have also been proposed as useful population-level measures with implications for prevention and urban planning [20]. A substantial body of research has examined the health effects of exposure to green spaces or green areas biodiversity. These studies have utilised a wide range of indicators (e.g., frequency of use, types of activities performed, time spent in urban green areas and Normalized Difference Vegetation Index (NDVI), plant or bird richness, respectively) and have been associated with multiple health outcomes. This has limited the comparability of existing evidence, as recently reviewed by Robinson and colleagues [21]. This variability underscores the need for a structured synthesis of current findings that distinguishes biodiversity from green-space exposure and examines their respective contributions to human health. Based on these considerations, this systematic review aimed to: (i) ascertain whether a relationship exists between urban green-space biodiversity and human health outcomes in adults aged 18 years and older; (ii) identify the parameters of biodiversity (number and type of plants, animals, microorganisms) most strongly associated with these outcomes; and (iii) determine whether specific health benefits, such as improvements in general and mental health, reductions in chronic disease burden, and maternal–fetal outcomes, are linked specifically to the biodiversity of green spaces, to green exposure more broadly, or to both. Methods This systematic review was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) statement [22] and followed a predefined protocol registered in the International Prospective Register of Systematic Reviews (PROSPERO; protocol number CRD42025636281). The review examines the relationship between biodiversity within urban green spaces and human health outcomes in adults aged 18 years and older, as well as the biodiversity indicators and health domains most frequently investigated in the literature. The completed PRISMA 2020 27-item checklist is provided in the supplementary materials. Research questions and PECO framework The primary research question guiding this review was: is there a relationship between urban green space biodiversity and human health outcomes in adults? Secondary aims explored which biodiversity dimensions (e.g., plant, animal, microbial) are most strongly associated with health outcomes and which mechanisms (e.g., exposure to microbiota, ecosystem services, or psychosocial benefits) mediate this relationship. The research was structured according to the PECO framework [23] to ensure conceptual clarity and methodological consistency. The population included adults (≥ 18 years) residing in urban or peri-urban areas. The exposure was the biodiversity level in urban green spaces, measured quantitatively or qualitatively through indices such as species richness, Shannon Index, vegetation diversity, or microbial composition. The comparator was lower or alternative levels of exposure to biodiversity in urban green spaces, where applicable. The outcomes comprised human health indicators, including morbidity, mortality, physiological, and psychological parameters. No restriction was applied to study design; cross-sectional, cohort, case-control, and experimental or quasi-experimental designs were all eligible. Information sources and search strategy Three bibliographic databases — PubMed (U.S. National Library of Medicine), Scopus (Elsevier), and Web of Science Core Collection (Clarivate; SCIE, SSCI, A&HCI, ESCI indexes) were searched, covering the period from January 2014 to July 2025. The last database search was run on 08 July 2025 across all three sources. Free-text terms were combined to capture studies addressing biodiversity, human health, and urban environments. The general search string applied across databases was (“biodiversity” AND “human health” AND “urban green space”), adapted to the field-tag syntax of each database. Full search strings for each database are provided in the supplementary materials. Reference lists of included articles and relevant reviews were also screened manually to identify additional studies. Only peer-reviewed publications written in English and available in full text with an abstract were considered eligible. Eligibility criteria, study selection, and data extraction Studies were eligible if they included adult (> 18 years old) human populations, provided a biodiversity measure in urban or peri-urban green spaces, and reported at least one health outcome (i.e. physiological, psychological, or disease-related parameters). Studies were excluded if they (i) lacked a biodiversity and/or green exposure measurement, (ii) lacked health outcomes, (iii) were conducted exclusively in rural or marine ecosystems, (iv) analysed allergic diseases or allergic sensitization as primary outcomes, or (v) were not published in English. No restrictions were applied with regard to study design or geographical area. Duplicate entries were identified using a two-step verification system. Initially, duplicates were manually identified and annotated. Subsequently, a secondary review was conducted using the Zotero software [24] for reference management and duplicate removal. Two reviewers (CA and CZ) independently screened titles, abstracts, and full texts according to the predefined criteria. Disagreements were resolved through discussion or, when necessary, adjudication by a third reviewer (NP). No automation tools were used at any stage of the screening process. The list of records excluded at the full-text stage, with reason of exclusion, is provided in the supplementary materials. Data extraction was performed independently by two reviewers (CA and CZ) using a structured form developed a priori; discrepancies in extracted data were resolved by discussion and, when necessary, adjudicated by a third reviewer (NP). Extracted variables included author, year of publication, country, study design, population characteristics, biodiversity indicators, type of urban green space, health outcomes, and main findings. All extracted data were cross-checked by all authors to ensure accuracy and completeness. Quality assessment and data synthesis Methodological quality and risk of bias were independently assessed by two reviewers (CA and NP) using the JBI Critical Appraisal Checklists [17] appropriate to each study design [18]. Each item was rated as Yes, No, Unclear, or Not applicable, and an overall score from 0 to 1 was calculated. Studies scoring > 0.80 were considered high quality, 0.60–0.79 moderate quality, and < 0.59 low quality. Inter-rater agreement between the two reviewers was evaluated by comparing the JBI appraisal outcomes assigned to each study independently prior to consensus discussion. For each of the 147 included studies, agreement was examined on the aggregated counts of items rated as Yes, No, Unclear, and Not applicable across the relevant JBI checklist (8 items for the Analytical Cross-sectional checklist, 11 items for the Cohort checklist, and 9 items for the Quasi-Experimental checklist). The two reviewers reached identical appraisal profiles in 134 of the 147 studies, corresponding to a study-level concordance of 91.2%. In the remaining 13 studies (8.8%), the reviewers’ counts differed by one to three items, with most discrepancies involving the classification of a single item as Yes versus Unclear. These differences typically concerned items that require subjective judgement, such as the appropriateness of confounder adjustment or the reliability of outcome measurement. Based on the observed distribution of disagreements across 1,287 individual appraisal items, the estimated item-level Cohen’s κ coefficient was approximately 0.78, corresponding to substantial agreement (full table reported in the supplementary materials). This value is consistent with inter-rater agreement levels reported in previous methodological studies that applied the JBI checklists in systematic reviews. All disagreements were resolved through structured discussion between the two reviewers; when consensus could not be reached, the appraisal was independently reassessed by a third investigator (CZ) to ensure final adjudication. Results were grouped according to biodiversity indicators, health domains, and population or geographic characteristics. Owing to methodological and outcome heterogeneity among the included studies, a meta-analysis was deemed inappropriate, and findings were synthesised narratively. Because of the heterogeneity of exposure metrics and outcome definitions across the included studies, effect measures varied widely. To allow a consistent summary across such heterogeneous measures, direction of association (positive, neutral, or negative) was extracted for each study and used as a summary indicator for the narrative synthesis. No formal subgroup analyses or sensitivity analyses were pre-specified; exploratory comparisons across outcome domains, biodiversity indicators, and population subgroups are presented descriptively in the results section. A formal assessment of publication or reporting bias was not feasible because no meta-analysis was performed; this aspect is discussed qualitatively among the limitations of the review. Similarly, the certainty of the body of evidence was not formally graded because of the narrative nature of the synthesis and the heterogeneity of exposure and outcome assessment; study-level methodological quality was instead appraised with the JBI checklists as described above. Results A total of 147 studies met the inclusion criteria and were included in the review. A flow diagram of the study selection procedure is reported in Fig. 1. The main characteristics, relevant data (such as study design, population and exposure) and results of each study were extracted and reported in the supplementary materials. Fig. 1PRISMA flow diagram PRISMA flow diagram The methodological quality of the included studies was high overall, with JBI scores ranging from 0.62 to 1.00 (median = 1.00). The large majority (93.2%, 137/147) were classified as high quality, while 10 studies (6.8%) reached a moderate-quality level; none fell into the low-quality category (Supplementary Figure S1). Methodological differences emerged between high and moderate quality studies. Among high-quality studies, 86.3% used objective greenness indicators derived from geographic information systems (GIS) or satellite data, and 71.9% specifically employed the normalized difference vegetation index (NDVI). Only one moderate-quality study used NDVI [25], while the remaining nine relied on alternative greenness metrics or did not report objective indices. Validated clinical or mental-health outcomes were reported in 82.7% of high-quality studies, whereas at least 90% of moderate quality studies (9/10) employed non-validated or short-term psychological measures. Adjustment for socioeconomic status was reported in 93.5% of high-quality studies. Environmental co-exposures such as air pollution or noise were considered in 43.1% of high-quality studies but in none of the moderate-quality studies. Sample sizes varied considerably, ranging from large national health databases or population-based surveys including tens of thousands of individuals to smaller studies with biological or questionnaire-derived measures (tens to hundreds of participants). Most studies were national-level investigations (106/147; 72%), while others were carried out in more than one country. Overall, 67 studies (45.6%) employed survey designs. In terms of study design, the majority were cross-sectional (71%), showing associations but not causal relationships, followed by cohort studies (24%) and experimental studies (5%). Chronological and geographical classification The temporal distribution of studies showed a clear increase in publication volume over time, covering the period from 2014 to 2025. In the early phase (2014–2016), 33 studies were published (22.4%); during the growth phase (2017–2019), 62 studies (42.2%); and in the most recent period (2020–2025), 52 studies (35.4%) (Fig. 2A). Fig. 2Chronological and geographical distribution of selected studies. A Number of studies according to the year of publication (* up to July 2025); (B) Countries involved in the studies and number of studies per country Chronological and geographical distribution of selected studies. A Number of studies according to the year of publication (* up to July 2025); (B) Countries involved in the studies and number of studies per country The peak year was 2019, with 23 publications (15.6%). A gradual decline was observed after 2020, with only three studies published in 2024 and none in 2025. Analysis by period also indicated shifts in research focus: mental health and well-being outcomes predominated in the early phase (21/33, 63.6%) but decreased proportionally in later periods (27/62, 43.5% in 2017–2019; 23/52, 44.2% in 2020–2024). Birth-related outcomes emerged primarily during 2017–2019 (8/62, 12.9%), while cardiovascular outcomes remained relatively stable across all phases (15–17%). The geographical distribution of studies showed a predominance of research conducted in Europe (n = 70), particularly in the United Kingdom (n = 22), followed by the United States (n = 22) and China (n = 19) (Fig. 2B). The geographical composition of the literature also changed over time. In 2014–2016, Europe accounted for 63.6% of studies (21/33) and Asia for only 9.1% (3/33). By 2020–2024, the European contribution had declined to 34.6% (18/52), while Asian output had risen to 30.8% (16/52). North American output remained stable at approximately 21% across all periods. Regional differences in outcome focus were observed: studies from the United Kingdom concentrated heavily on mental health and well-being (20/25, 80.0%), US studies showed a more diversified profile including mental health (9/22, 40.9%), birth outcomes (5/22, 22.7%), and cardiovascular outcomes (4/22, 18.2%), while Chinese studies placed relatively greater emphasis on cardiovascular outcomes (5/19, 26.3%). Studies examining medication consumption (e.g., antidepressant prescriptions) were almost exclusively European (8/8, 100%). Assessment of biodiversity and green exposure Our results showed that 66.1% of the studies (97/147) reported a measure of green exposure to evaluate different health effects, mainly using plants or trees as biodiversity markers. Some studies assessed tree richness (i.e., the number of different tree species), while others measured canopy tree cover using GIS or satellite-derived data. Additional studies quantified exposure through the percentage of greenery or the overall greenness of an area. Fifty studies (34.0%) used the normalized difference vegetation index (NDVI), a widely used metric for quantifying vegetation health and density from sensor data, either alone or combined with other parameters such as percentage of green space or land use. Three studies (4.1%) employed the Shannon Index, an ecological measure of species diversity that captures both richness and evenness. A total of 28 studies (22.4%) used the percentage of green area (including trees, grass, and other land-cover types). No clear relationship was observed between the type of plant biodiversity index used and either study design or the specific disease outcomes investigated. Only six studies (4.0%) reported a biodiversity index that did not consider vegetation but instead measured bird or insect richness [26–31], while another 7 studies (4.7%) included both plant and bird/insect richness as part of their index [32–38]. Finally, almost a fifth of the selected studies (28/147, 18.4%) did not report any objective biodiversity index. Instead, they focused on indicators such as frequency of use, types of activities performed, and time spent in urban green areas. Exposure was typically measured in terms of time spent in green spaces (ranging from 15 min to 5 h) [39–41], frequency of visits (e.g., seldom, monthly or weekly, daily) [41–43], and type of activities undertaken (e.g., walking, sports) [39, 44]. Classification of health outcomes Among the selected studies, four main thematic clusters were identified on the basis of the health outcomes analysed: (i) Outcomes related to general health, mental health, and well-being; (ii) Outcomes related to the consumption of medications primarily used to treat mental health conditions; (iii) Outcomes related to chronic diseases (cardiovascular diseases, respiratory diseases, diabetes, obesity); (iv) Birth-related outcomes, such as birth weight, prematurity, and other indicators of neonatal health correlated with maternal exposures. Studies on general health, mental health, and well-being outcomes The correlation between exposure to urban green areas and general or mental health and well-being was the most frequently investigated topic, with 78 out of 147 studies (53.1%) addressing these issues. Among these, 20 studies (25.6%) examined general health and the frequency or aptitude for physical activity, which indirectly exerts positive effects on health; 21 studies (26.9%) focused specifically on mental health or depression, and 37 (47.4%) on well-being, stress reduction, and relaxation. Only one study (1.2%) evaluated both well-being and drug prescriptions (anxiolytics, hypnotics, and antidepressants) as outcomes using data from a health registry [45] (Table 1). Most studies analysed self-reported outcome data. Only a small minority of general health studies (2/20, 10%) used assessment scales, standardized questionnaires, or measurable biological parameters. In contrast, 33% (7/21) and 27% (10/37) of mental health and well-being studies, respectively, employed validated assessment scales for different outcomes. Depressive symptoms or disorders were the most frequently assessed mental health outcomes (50%, 10/21) [33, 43, 61–65, 71–73], followed by anxiety (25%, 5/21) [33, 61, 66, 71, 74], stress (15%, 3/21) [32, 61, 71], and common self-reported mental health indicators. Other outcomes included life satisfaction, vitality, mindfulness, and resilience [41, 76–78, 96]. Instruments used to measure general and mental health included validated questionnaires such as the General Health Questionnaire (GHQ-12/GHQ-28) [46, 47, 73, 75, 97], the Patient Health Questionnaire (PHQ-2/PHQ-9) [71, 74], the Depression, Anxiety, and Stress Scales (DASS-21) [33, 61, 72], and the Kessler Psychological Distress Scale (K10) [45, 67, 98]. Across studies, two main approaches were used to assess exposure to green areas: (1) residential exposure, defined as the distance between home and green areas, and (2) active use, defined as activities carried out in green areas or frequency of visits. Studies on well-being were nearly equally divided in terms of exposure type, with 53% using residence-based measures, whereas most studies on general and mental health used residence-based exposure (81% and 76%, respectively).Table 1Main characteristics of studies on general and mental health and well-beingHealth outcomesGreen exposureBiodiversity measurementTrend of evidence of studies with green exposureTrend of evidence of studies with measured health dataTrend of evidence of studies with biodiversitySelf-reported dataMeasured data#ResidenceUseGeneral health[25, 27, 30, 36, 46–58][59, 60][27, 30, 46–52, 54, 56–60][25, 36, 53, 55][27, 30, 36, 47]70.0% ▲25.0% ↔5% ▼100.0% ▲50.0% ▲50.0% ↔Mental health[31, 35, 42, 43, 61–70][32, 33, 71–75][32, 33, 35, 43, 62–67, 69–71, 73–75][31, 42, 43, 68][31–33, 35]90.5% ▲4.8% ↔4.8% ▼100.0% ▲75.0% ▲25.0% ↔Well-being/stress reduction[12, 26, 29, 34, 37, 38, 41, 76–95][28, 39, 96–103][29, 38, 77–79, 81, 83, 86, 87, 89, 93, 94, 98, 100, 102, 103][12, 26, 28, 34, 37, 39, 41, 76, 80, 82, 84, 85, 88, 90–92, 95–97, 99, 101][12, 26, 28, 29, 34, 37, 84, 99]91.9% ▲2.7% ↔5.4% ▼90.0% ▲10.0% ▼87.5.% ▲12.5% ↔Drug prescription-[45][45]--100.0% ▲100.0% ▲#By means of: scale for evaluation, questionnaire, biological parameters; ▲: positive correlation; ↔: uncertain correlation; ▼: negative correlation Table 1Main characteristics of studies on general and mental health and well-beingHealth outcomesGreen exposureBiodiversity measurementTrend of evidence of studies with green exposureTrend of evidence of studies with measured health dataTrend of evidence of studies with biodiversitySelf-reported dataMeasured data#ResidenceUseGeneral health[25, 27, 30, 36, 46–58][59, 60][27, 30, 46–52, 54, 56–60][25, 36, 53, 55][27, 30, 36, 47]70.0% ▲25.0% ↔5% ▼100.0% ▲50.0% ▲50.0% ↔Mental health[31, 35, 42, 43, 61–70][32, 33, 71–75][32, 33, 35, 43, 62–67, 69–71, 73–75][31, 42, 43, 68][31–33, 35]90.5% ▲4.8% ↔4.8% ▼100.0% ▲75.0% ▲25.0% ↔Well-being/stress reduction[12, 26, 29, 34, 37, 38, 41, 76–95][28, 39, 96–103][29, 38, 77–79, 81, 83, 86, 87, 89, 93, 94, 98, 100, 102, 103][12, 26, 28, 34, 37, 39, 41, 76, 80, 82, 84, 85, 88, 90–92, 95–97, 99, 101][12, 26, 28, 29, 34, 37, 84, 99]91.9% ▲2.7% ↔5.4% ▼90.0% ▲10.0% ▼87.5.% ▲12.5% ↔Drug prescription-[45][45]--100.0% ▲100.0% ▲#By means of: scale for evaluation, questionnaire, biological parameters; ▲: positive correlation; ↔: uncertain correlation; ▼: negative correlation Main characteristics of studies on general and mental health and well-being 70.0% ▲ 25.0% ↔ 5% ▼ 50.0% ▲ 50.0% ↔ 90.5% ▲ 4.8% ↔ 4.8% ▼ 75.0% ▲ 25.0% ↔ Well-being /stress reduction 91.9% ▲ 2.7% ↔ 5.4% ▼ 90.0% ▲ 10.0% ▼ 87.5.% ▲ 12.5% ↔ #By means of: scale for evaluation, questionnaire, biological parameters; ▲: positive correlation; ↔: uncertain correlation; ▼: negative correlation Most studies (58.9%, 46/78) controlled for socio-economic status (SES) when adjusting results. Many found that the positive association between green exposure (residence or use) and improved mental health or well-being was stronger in populations with low SES levels. Overall, 70.0% of general health studies reported positive findings, showing a correlation between residential exposure or active engagement with urban green spaces and improved self-reported health (poor/fair vs. good/very good/excellent) [46–53, 104]. Greater tree canopy, NDVI, garden size, and proximity to parks were consistently associated with better general health outcomes [48, 51, 52], whereas grass alone showed weaker or null associations [52]. Smaller green areas and more distant or fragmented green spaces were linked to poorer health and increased health inequalities [47–49, 53]. Five studies (25%) showed no clear evidence, as low SES significantly influenced the results: residential greenness effects were often stronger among low-SES populations, older adults, and individuals with poorer baseline health [36, 47, 53, 54, 104]. Four studies have specifically examined whether biodiversity provides added value beyond mere exposure to green spaces (Table 1). In these studies, exposure to urban green spaces was characterised not only by availability, but also by biodiversity and ecological quality. The assessment of biodiversity was conducted using both objective indicators, including species richness and environmental quality parameters [27, 30, 36], and subjective perceptions [47], encompassing perceived species richness and visual stimuli. In particular, the studies found a correlation between self-reported well-being and higher levels of biodiversity or perceived naturalness. The results of these studies suggest that the existence and perception of biodiversity are significantly associated with health and well-being benefits derived from urban green spaces. However, the general trend of the evidence showed that approximately half of the studies revealed a positive correlation, while the others produced uncertain results [27, 47]. The 21 studies that examined the relationship between green-space exposure and mental health were conducted mainly in high-income countries, with the United Kingdom (n = 7) and Canada (n = 3) most frequently represented. A total of 90.5% of these studies (19/21) reported positive findings, showing a correlation between exposure to urban green spaces and improved mental health, including reductions in depression and anxiety levels [33, 61, 64]. Alcock and colleagues (2014) carried out a longitudinal cohort study and their results suggested that moving to greener areas led to improvements in mental health, whereas relocation to areas with lower greenness was associated with temporary declines [75]. Green-space use was also positively associated with mental health: increased time spent in parks correlated with lower depressive symptoms, enhanced social cohesion, and greater vitality [42, 68, 69, 72]. Biodiversity measures, including bird and tree richness, were linked to better mental health outcomes. Stronger effects were observed among populations with lower socioeconomic status (SES) [32, 33], similarly to the way in which the benefits of green exposure were modified by socio-demographic characteristics, urbanicity, and lifestyle factors. Positive effects of green exposure were more pronounced among women, younger adults, individuals with lower SES, and those engaging in higher levels of physical activity [64, 73]. Moreover, childhood experiences with nature moderated the association between green-space visits and mental health [42]. Xu and colleagues [31] found no relationship between mountain park usage, bird biodiversity and momentary mental health, while Pelgrims and coworkers [70] observed that air pollution, rather than greenness, was the primary determinant of mental health outcomes. Some studies further noted that residential green exposure effects were weaker for anxiety and mood disorders compared to depressive symptoms [66, 71]. Among studies investigating well-being, outcomes spanned multiple dimensions, including stress [38, 79, 80, 98], psychological distress [81, 97, 99, 100], life satisfaction, and immediate affective or restorative responses such as positive or negative mood, emotional response [76], and perceived restorativeness [37, 82, 101]. A total of 91.9% of well-being studies reported positive correlation with green exposure, 5.4% reported negative correlation and 2.7% found no clear association between green exposure and well-being. Similar results were obtained in relation to the biodiversity of green spaces (87.5% were positive and 12.5% had no clear correlation). Higher tree canopy or tree richness was consistently associated with lower psychological distress, reduced depression risk, and better self-rated health [83, 98, 100, 102]. Frequent visits to green spaces were associated with improved mood, higher life satisfaction, and reduced stress [29, 37, 101, 102]. Benefits were observed across various types of use, including walking, social activities, and relaxation. In several studies, the perceived biodiversity and naturalness of green areas, rather than the measured biodiversity of trees, insects, or birds, emerged as stronger predictors of positive correlation with well-being and restorativeness [26, 29, 38, 84]. Park access within 300–500 m buffers was also linked to improved self-rated health and life satisfaction, although effects varied according to vegetation type and park characteristics [78]. Only 2 studies reported null associations between well-being and green exposure [38, 80], while only one study showed a negative correlation with the biodiversity index (bird richness) [28]. Among the studies that used outcome measures (measurement scales or physiological parameters), all showed a positive correlation between green exposure and mental and general health. In contrast, positivity is at 90% for studies on well-being and stress reduction. Among the quasi-experimental and experimental studies [36, 38, 44, 82, 101, 105] positive correlations between green exposure (e.g. visits to urban forests and parks) and improved psychophysical outcomes were evidenced. These include improved mood and restorativeness [36, 82, 101], reduced cortisol levels [44] and lower blood pressure and improved heart rate variability [38, 44, 105]. Perceived biodiversity appears to be positively linked to well-being, while objective measures yield less consistent results (bird richness and Shannon Index) [36, 38]. A separate analysis was conducted for studies reporting outcomes related to the consumption of medications primarily used to treat mental health conditions (Table 2). All included studies reported a positive association between green exposure and mental health, and, in particular, an inverse relationship between urban greenery and the use of psychotropic drugs (e.g., antidepressants, anxiolytics, and antipsychotics). Aerts and colleagues [106] found that greater green-space availability, as measured by NDVI and the types of green space present (e.g., woods, meadows, and gardens), was associated with a significant reduction in psychotropic drug sales, particularly in urban and socioeconomically disadvantaged areas. Chi and colleagues [107] reported similar results, observing that tree-canopy density and volume were inversely correlated with the consumption of mood-altering and cardiovascular drugs. Gascon and coworkers [108] also found that higher levels of vegetation, measured by NDVI at different spatial buffers, were associated with a lower likelihood of depression and benzodiazepine use. However, no significant effects were observed in relation to proximity to blue spaces (e.g., water bodies). Helbich and colleagues [109] highlighted a non-linear relationship between the percentage of green spaces and the rate of antidepressant prescriptions, suggesting the existence of a threshold and a dose-dependent response: the protective effect was more pronounced in areas with high baseline prescription rates. Similarly, Marselle and coworkers [20] found that a higher density of trees along roads within 100 m of subjects’ homes was associated with a reduction in antidepressant prescriptions, with stronger effects observed among individuals with low socioeconomic status. However, tree biodiversity (species richness) and exposure at greater distances did not show significant associations. Using ecological data from 31 London boroughs, Taylor and coworkers [110] reported a slight reduction in antidepressant prescription rates associated with roadside tree density, although the effect was borderline significant. Finally, Triguero-Mas and colleagues [111] found that higher NDVI values within 300 m of home were consistently associated with better self-perceived health and lower levels of psychological distress. In this case, neither physical activity nor social support appeared to mediate the observed effect, while exposure to blue spaces again showed no significant associations. Table 2Urban green space exposure and consumption of medications primarily used to treat mental health conditionsCountry/CityPopulationGreen exposureOutcomeMain findingReferencesBelgium (national study)Adult populationNDVI + green-space types (woodland, grassland, gardens)Psychotropic medication salesMore green space linked to lower medication sales; strongest in urban/deprived areas [106]Belgium/BrusselsAdults(19–64 years)Residential exposure to urban treesMedication sales (mood, CVD)Higher tree density and crown volume linked to lower drug sales [107]Spain/Barcelona958 adults(45–74 years)NDVI (100–500 m); land cover; access to major green/blue spaceAnxiety, depression, psychotropic useHigher greenness linked to lower depression and benzodiazepine use; blue space null [108]Netherlands (national study, 403 municipalities)Municipal populations% green spaces (parks, agricultural, forests)Antidepressant prescription ratesNon-linear inverse relationship; threshold/dose-response pattern [109]Germany/Leipzig9,751 adultsStreet-tree density (100–1000 m buffers) species richnessAntidepressant prescriptionsHigher tree density within 100 m linked to lower drugs prescriptions, especially for low SES; species richness is not significant [20]UK/London(31 boroughs)Area-level dataTree density per km streetAntidepressant prescribing rateMore street trees are slightly linked to lower drug prescribing [110]Spain/Catalonia8,793 adultsNDVI 300 m; access to green/blue spacesPsychotropic use, visit to mental health specialistsHigher greenness is linked to drugs intake (antidepressant, sedatives) [111] Table 2Urban green space exposure and consumption of medications primarily used to treat mental health conditionsCountry/CityPopulationGreen exposureOutcomeMain findingReferencesBelgium (national study)Adult populationNDVI + green-space types (woodland, grassland, gardens)Psychotropic medication salesMore green space linked to lower medication sales; strongest in urban/deprived areas [106]Belgium/BrusselsAdults(19–64 years)Residential exposure to urban treesMedication sales (mood, CVD)Higher tree density and crown volume linked to lower drug sales [107]Spain/Barcelona958 adults(45–74 years)NDVI (100–500 m); land cover; access to major green/blue spaceAnxiety, depression, psychotropic useHigher greenness linked to lower depression and benzodiazepine use; blue space null [108]Netherlands (national study, 403 municipalities)Municipal populations% green spaces (parks, agricultural, forests)Antidepressant prescription ratesNon-linear inverse relationship; threshold/dose-response pattern [109]Germany/Leipzig9,751 adultsStreet-tree density (100–1000 m buffers) species richnessAntidepressant prescriptionsHigher tree density within 100 m linked to lower drugs prescriptions, especially for low SES; species richness is not significant [20]UK/London(31 boroughs)Area-level dataTree density per km streetAntidepressant prescribing rateMore street trees are slightly linked to lower drug prescribing [110]Spain/Catalonia8,793 adultsNDVI 300 m; access to green/blue spacesPsychotropic use, visit to mental health specialistsHigher greenness is linked to drugs intake (antidepressant, sedatives) [111] Urban green space exposure and consumption of medications primarily used to treat mental health conditions Adults (19–64 years) 958 adults (45–74 years) UK/London (31 boroughs) Overall, the results indicate that green exposure, particularly that characterised by a high tree canopy and abundant vegetation, was associated with improved mental health, as evidenced by both objective indicators (e.g., medication sales or prescription rates) and subjective measures (e.g., self-perceived health). All the examined studies related to the consumption of medications primarily used to treat mental health conditions were cross-sectional and conducted in Europe, showing consistent findings across different geographical contexts and reinforcing the hypothesis of a protective effect of natural environments on mental health. Studies on chronic diseases: diabetes, cardiovascular diseases, obesity Chronic degenerative diseases were found to be associated with green exposure in 44 out of 147 studies (30.0%). Among these, 18 studies (40.9%) assessed both all-cause mortality and cause-specific mortality, particularly for cardiovascular and respiratory diseases [60, 112–128]. Two additional studies (4.5%) used life expectancy [129] and years of life lost [130] as outcome parameters. The presence of green spaces near the home was consistently associated with a reduction in cardiovascular mortality, not only in small cities but also in highly urbanised settings, suggesting that even access to small green areas can have beneficial effects. Silveira and colleagues [123] showed that greater green coverage was associated with lower mortality from cardiovascular diseases, including heart attacks and cerebrovascular events, regardless of SES. Aerts and coworkers [112] reported similar findings, showing that the beneficial effects of green exposure were more pronounced in low-SES areas, indicating that access to and use of green urban areas may have a greater impact on socially vulnerable populations. All studies adjusted their analyses for the main risk factors associated with chronic degenerative diseases, such as age, gender, BMI (body mass index), smoking status, educational attainment, and SES. Several studies, also reporting levels of major atmospheric pollutants (e.g., PM₁₀, PM₂.₅, NOₓ) [60, 114, 116, 118, 121, 122], showed that the protective effects of green exposure were partly mediated by air pollution. Generally, the studies measured exposure and outcome at the same time, so they cannot establish temporality (i.e., whether exposure preceded disease). This could create an exposure–outcome mismatch, especially for chronic conditions that develop over years. Overall, these studies consistently demonstrated a positive correlation between green exposure and a reduction in both all-cause and cause-specific mortality, as well as an improvement in life expectancy (Table 3). Table 3Summary of studies on chronic diseases: diabetes, cardiovascular diseases, obesityHealth OutcomesExposure effectGreen exposure with strongest effectsProposed MechanismsNoteReferencesAll-cause mortalityProtectiveNDVI (range 500–1250 m), tree canopy, % of green↓ air pollution,↓ noise,↑ PA,↓ stress2 studies [114, 120] had null or limited evidence [60, 112–115, 118–122, 124, 128]CVD mortalityProtectiveTree canopy, NDVI, street-tree biodiversity↓ air pollutants,↑ ecosystem health,↓ heatSome models attenuated after full SES adjustment [60, 113, 114, 116–118, 120, 122–127]CVD incidenceProtectiveNDVI, % green space, park proximity↓ BP,↓ pollution,↑ PA, stress recoverySome variability by sex (women stronger in some studies) [25, 72, 131–135]Hospitalizations for CVD and respiratory diseaseProtective% of green↓ Risk and hospitalization ratesLimited number of studies overall [132, 136]Hypertension and blood pressureProtectiveNDVI, street greenery, tree count↓ stress,↓ noise,↑ autonomic balanceVisual greenness more predictive than satellite greenness [105, 131, 135, 137–144]Diabetes/Metabolic outcomesProtectiveNDVI,% green space, tree canopy↑ Weight control,↑ PA,↓ pollutionStronger effects in deprived areas; sometimes sex-specific [131, 138–140, 145–147]Obesity/BMIProtectiveTree canopy, street-level greenery↑ walkability,↑ PACity-level: NDVI less predictive; stronger effects in some studies in women. A study [137] had null or limited evidence [137–140, 148–150]BMI Body mass index, BP Blood pressure, CVD Cardiovascular diseases, NDVI Normalized difference vegetation index, PA Physical activity, ↑: increase; ↓: decrease Table 3Summary of studies on chronic diseases: diabetes, cardiovascular diseases, obesityHealth OutcomesExposure effectGreen exposure with strongest effectsProposed MechanismsNoteReferencesAll-cause mortalityProtectiveNDVI (range 500–1250 m), tree canopy, % of green↓ air pollution,↓ noise,↑ PA,↓ stress2 studies [114, 120] had null or limited evidence [60, 112–115, 118–122, 124, 128]CVD mortalityProtectiveTree canopy, NDVI, street-tree biodiversity↓ air pollutants,↑ ecosystem health,↓ heatSome models attenuated after full SES adjustment [60, 113, 114, 116–118, 120, 122–127]CVD incidenceProtectiveNDVI, % green space, park proximity↓ BP,↓ pollution,↑ PA, stress recoverySome variability by sex (women stronger in some studies) [25, 72, 131–135]Hospitalizations for CVD and respiratory diseaseProtective% of green↓ Risk and hospitalization ratesLimited number of studies overall [132, 136]Hypertension and blood pressureProtectiveNDVI, street greenery, tree count↓ stress,↓ noise,↑ autonomic balanceVisual greenness more predictive than satellite greenness [105, 131, 135, 137–144]Diabetes/Metabolic outcomesProtectiveNDVI,% green space, tree canopy↑ Weight control,↑ PA,↓ pollutionStronger effects in deprived areas; sometimes sex-specific [131, 138–140, 145–147]Obesity/BMIProtectiveTree canopy, street-level greenery↑ walkability,↑ PACity-level: NDVI less predictive; stronger effects in some studies in women. A study [137] had null or limited evidence [137–140, 148–150]BMI Body mass index, BP Blood pressure, CVD Cardiovascular diseases, NDVI Normalized difference vegetation index, PA Physical activity, ↑: increase; ↓: decrease Summary of studies on chronic diseases: diabetes, cardiovascular diseases, obesity ↓ air pollution, ↓ noise, ↑ PA, ↓ stress ↓ air pollutants, ↑ ecosystem health, ↓ heat ↓ BP, ↓ pollution, ↑ PA, stress recovery ↓ stress, ↓ noise, ↑ autonomic balance NDVI, % green space, tree canopy ↑ Weight control, ↑ PA, ↓ pollution ↑ walkability, ↑ PA BMI Body mass index, BP Blood pressure, CVD Cardiovascular diseases, NDVI Normalized difference vegetation index, PA Physical activity, ↑: increase; ↓: decrease However, Klompmaker and colleagues [120] found no such association, while two studies [114, 137] reported weak or inconsistent results, with no clear linear correlation. Other studies investigated the incidence and prevalence of cardiovascular and respiratory diseases, T2 diabetes, and overweight or obesity in populations exposed to varying levels of green exposure, both for individual diseases and for multiple conditions combined. Three studies examined the association with diabetes [145–147]. Astell-Burt and coworkers [145] demonstrated that having green spaces near the home was linked to a lower risk of type 2 diabetes. Green environments can encourage physical activity, alleviate stress, and improve air quality, thereby positively influencing metabolic health [131, 138–140, 146]. Urban green areas promote behaviours such as walking, running, and outdoor exercise, which directly improve physical health and reduce the risk of chronic conditions, including diabetes, heart diseases, and obesity. The incidence and prevalence of cardiovascular diseases represented the most frequently studied topic (27.9%) in relation to green exposure [25, 72, 105, 132–134, 136, 137, 141–144]. All studies, except Picavet and colleagues [137], who reported weak or inconsistent evidence of a correlation between green areas (gardens/agricultural land) and blood pressure or BMI, showed positive associations, with reductions in cardiovascular diseases and its risk factors. Most studies (n = 16, 36.3%) used the NDVI derived from satellite imagery as the primary exposure metric, providing objective and comparable measurements across locations. Only one study used the frequency of visits to green areas and the amount of physical activity carried out in parks as parameters [105]; two studies evaluated both residence and frequency of use [133, 139]; and the remainder used residential proximity alone as the green exposure measure. Almost all studies adjusted for major disease confounders, and in addition to general variables such as age, gender, and education, several considered specific risk factors (e.g., smoking, BMI, obesity, blood glucose, cholesterol). Particular importance was given to SES and physical activity in the analytical models. Studies on birth-related outcomes It is interesting to note that specific health outcomes investigated in the context of the association between green exposure and biodiversity include birth-related outcomes, such as birth weight, prematurity, and other indicators of neonatal health, including head circumference, foetal growth, and small for gestational age (SGA) status (Table 4). Thirteen studies have been conducted, mainly in North America (53.8%, including six from the United States and one from Canada) and 23.1% in Europe. Cohort studies predominated (84.6%), with only two cross-sectional studies. The included studies enrolled pregnant women at the time of childbirth and comprised population samples ranging from a few hundred participants [152] to several million individuals [155]. Green exposure was primarily measured using a single parameter, the mother’s residential address (84.6%, 11 out of 13 studies), and the distance from parks, with areal ranges typically between 100 and 500 m. Only two studies also assessed the use and frequency of visits to green areas [151, 162]. The NDVI was used in all studies as the main indicator of biodiversity richness. More than half of the studies (53.8%, 7/13) reported a positive correlation between green exposure and birth outcomes. Table 4Main results of maternal and fetal health outcomesCountryGreen exposureSES level adjustmentBirth outcomesMain findingReferencesUSA (NYC)Street tree density, NDVI, access to large greenspaces/wateryesBW, LBW, PTB, SGAStreet trees are linked with ↓PTB. NDVI and park/water access results inconsistent. [151]SpainNDVI (300 m), green space > 5000 m² within 300–500 mnoBW, LBW, PTB SGANo robust associations, except some evidence green space (300 m) associated with ↓ SGA. [152]USA (Pennsylvania)NDVI (250 m)yesBW, PTB, SGACities: Green exposure is linked ↓ PTB and ↓ SGA.Non-urban areas: no correlation. [153]USA(New England)Tree exposure (100 m–3 km), NDVI, land cover diversitynoBW, HC, FLGreen exposure associated with ↑ HC and FL, no correlation with BW. [154]USA (Texas)NDVI (250 m)yesBW, PTB, SGAUnadjusted analysis: protective.Fully adjusted: most correlations disappear. Some subgroup correlations remain. [155]CanadaNDVI (500 m), parksyesBWNo overall correlations, but positive association in dense/urban cities. [156]Italy/Austria (Alpine)NDVI (500 m)noBW, LBW, PTB, SGAResults inconsistent: air pollution levels make difference: only in Austria NDVI is linked to ↓ LBW/SGA. [157]USA (Massachusetts)NDVI (250 m)yesBW, TLBW, SGARobust positive associations: ↑BW, ↓TLBW/SGA. Stronger for higher SES and urban areas. [158]USA (California)NDVI (500 m),distance to parksyesLBWNDVI is associated with ↓LBW risk. ~12% mediated by PM2.5. Park proximity not a strong modifier. [159]China (Beijing)NDVI (200 m, 500 m)yesFG, LBW, SGANDVI (500 m) ↑FG measures, no correlation with BW or LBW/SGA. [160]New Zealand% green space (area-level)yesBW, SGANo overall correlations, % of green is linked with ↓SGA for women with lowest education. [161]Europe(6 countries)NDVI (100–500 m),access to green/blue spacesyesBW, TLBWConsistent across cities: ↑ NDVI ↑ BW; ↓ TLBW.Large, robust correlation across cohorts. [162]AustraliaGreenness (foliage cover)yesBW, SGA, LGAGreen exposure is linked with ↑BW.Biodiversity not associated with outcomes. [163]BW Birth weight, FG Fetal growth, FL Fetal length, LBW Low birth weight, HC Head circumference, LGA Large for gestational age, PTB Preterm birth, SGA Small gestational age, TLBW Term low birth weight, ↑: increase; ↓: decrease Table 4Main results of maternal and fetal health outcomesCountryGreen exposureSES level adjustmentBirth outcomesMain findingReferencesUSA (NYC)Street tree density, NDVI, access to large greenspaces/wateryesBW, LBW, PTB, SGAStreet trees are linked with ↓PTB. NDVI and park/water access results inconsistent. [151]SpainNDVI (300 m), green space > 5000 m² within 300–500 mnoBW, LBW, PTB SGANo robust associations, except some evidence green space (300 m) associated with ↓ SGA. [152]USA (Pennsylvania)NDVI (250 m)yesBW, PTB, SGACities: Green exposure is linked ↓ PTB and ↓ SGA.Non-urban areas: no correlation. [153]USA(New England)Tree exposure (100 m–3 km), NDVI, land cover diversitynoBW, HC, FLGreen exposure associated with ↑ HC and FL, no correlation with BW. [154]USA (Texas)NDVI (250 m)yesBW, PTB, SGAUnadjusted analysis: protective.Fully adjusted: most correlations disappear. Some subgroup correlations remain. [155]CanadaNDVI (500 m), parksyesBWNo overall correlations, but positive association in dense/urban cities. [156]Italy/Austria (Alpine)NDVI (500 m)noBW, LBW, PTB, SGAResults inconsistent: air pollution levels make difference: only in Austria NDVI is linked to ↓ LBW/SGA. [157]USA (Massachusetts)NDVI (250 m)yesBW, TLBW, SGARobust positive associations: ↑BW, ↓TLBW/SGA. Stronger for higher SES and urban areas. [158]USA (California)NDVI (500 m),distance to parksyesLBWNDVI is associated with ↓LBW risk. ~12% mediated by PM2.5. Park proximity not a strong modifier. [159]China (Beijing)NDVI (200 m, 500 m)yesFG, LBW, SGANDVI (500 m) ↑FG measures, no correlation with BW or LBW/SGA. [160]New Zealand% green space (area-level)yesBW, SGANo overall correlations, % of green is linked with ↓SGA for women with lowest education. [161]Europe(6 countries)NDVI (100–500 m),access to green/blue spacesyesBW, TLBWConsistent across cities: ↑ NDVI ↑ BW; ↓ TLBW.Large, robust correlation across cohorts. [162]AustraliaGreenness (foliage cover)yesBW, SGA, LGAGreen exposure is linked with ↑BW.Biodiversity not associated with outcomes. [163]BW Birth weight, FG Fetal growth, FL Fetal length, LBW Low birth weight, HC Head circumference, LGA Large for gestational age, PTB Preterm birth, SGA Small gestational age, TLBW Term low birth weight, ↑: increase; ↓: decrease Main results of maternal and fetal health outcomes Cities: Green exposure is linked ↓ PTB and ↓ SGA. Non-urban areas: no correlation. USA (New England) Unadjusted analysis: protective. Fully adjusted: most correlations disappear. Some subgroup correlations remain. NDVI (500 m), distance to parks Europe (6 countries) NDVI (100–500 m), access to green/blue spaces Consistent across cities: ↑ NDVI ↑ BW; ↓ TLBW. Large, robust correlation across cohorts. Green exposure is linked with ↑BW. Biodiversity not associated with outcomes. BW Birth weight, FG Fetal growth, FL Fetal length, LBW Low birth weight, HC Head circumference, LGA Large for gestational age, PTB Preterm birth, SGA Small gestational age, TLBW Term low birth weight, ↑: increase; ↓: decrease Six studies [151, 153, 157–159, 163] found significant associations between increased green exposure and a reduced risk of low birth weight and premature birth. In some cases, the correlation appeared to be stronger in high-density urban areas [153] or among more advantaged socioeconomic groups [158]. Only one study accounted for PM₂.₅ pollution, emphasising that the positive correlation of green exposure was not influenced by air pollution levels [68]. However, 38.5% (5/13) of studies [152, 154, 156, 160, 162] did not find a significant association between green exposure and birth outcomes. Anabitarte and colleagues (2020) [152] did not report a significant correlation between green areas and pregnancy outcomes, except within 300 m of the mother’s residence, where a reduction in SGA risk was observed. Some studies did not show a reduction in the risk of low birth weight but did identify correlations with head circumference [154] or foetal growth [160]. Conversely, the European multicentre study [162] found an association only with a reduced risk of low birth weight and not with other parameters, suggesting that other urban factors may have a greater impact on these outcomes. Some authors also proposed that the effect might be limited to specific subgroups of pregnant women, particularly those defined by socioeconomic status or educational level. However, no significant associations were observed in the overall sample [161]. All studies in this domain adjusted their analyses for the main maternal socio-demographic characteristics (e.g., education, maternal age, ethnicity, BMI, smoking, and alcohol consumption) and for socioeconomic variables (e.g., deprivation index of the residential area), confirming that these factors significantly influenced the results. The results obtained were summarized in Table 5 in order to provide a concise overview. Table 5Summary of the main findings of the literature reviewOutcomesEvidence TrendBiodiversity IndexHealth benefitModifying FactorsGeneral Health70% positiveTree canopy, NDVIBetter self-rated healthSES, age, baseline healthMental Health90% positiveTree & bird richness, NDVI↓Depression,↓Anxiety,↑VitalitySES, gender, activity, childhood nature contactWell-being90% positiveTree richness, perceived biodiversity↓Stress,↑Life satisfactionUrban characteristics, park accessDrug Use100% inverseTree canopy, NDVI↓Psychotropic prescriptionsSES, urban areasChronic Diseases95% positiveNDVI, vegetation cover↓Mortality,↓CVD risk,↓DiabetesAir pollution, SESBirth Outcomes~ 54% positiveNDVI near residence↑Birth weight,Preterm riskSES, urban density Table 5Summary of the main findings of the literature reviewOutcomesEvidence TrendBiodiversity IndexHealth benefitModifying FactorsGeneral Health70% positiveTree canopy, NDVIBetter self-rated healthSES, age, baseline healthMental Health90% positiveTree & bird richness, NDVI↓Depression,↓Anxiety,↑VitalitySES, gender, activity, childhood nature contactWell-being90% positiveTree richness, perceived biodiversity↓Stress,↑Life satisfactionUrban characteristics, park accessDrug Use100% inverseTree canopy, NDVI↓Psychotropic prescriptionsSES, urban areasChronic Diseases95% positiveNDVI, vegetation cover↓Mortality,↓CVD risk,↓DiabetesAir pollution, SESBirth Outcomes~ 54% positiveNDVI near residence↑Birth weight,Preterm riskSES, urban density Summary of the main findings of the literature review ↓Depression, ↓Anxiety, ↑Vitality ↓Stress, ↑Life satisfaction ↓Mortality, ↓CVD risk, ↓Diabetes ↑Birth weight, Preterm risk Discussion A systematic analysis of the literature reveals a consistent, positive correlation between green exposure and improvements in human health. Most studies assessed green exposure using plant-based indicators, such as tree canopy cover and satellite-derived indices (e.g., NDVI), which were consistently linked to better self-reported health and a reduced incidence of many chronic-disease outcomes. These findings are consistent with the extant literature, which supports the health-promoting role of urban green exposure. Nevertheless, the role of biodiversity in urban green spaces remains ambiguous and understudied. A mere 12.9% of the selected studies (19 out of 147) reported an objective measurement of biodiversity. However, inconsistency in the employed parameters and their associations with health outcomes generates ambiguity. Despite this concern, the available evidence suggests that biodiversity may represent an additional and more sensitive dimension of green exposure. Studies incorporating indicators such as species richness or ecological diversity have consistently reported stronger or more specific associations with health outcomes [117], particularly in the context of mental health and wellbeing [12, 26, 32]. In this review over 75% of the studies examined assessed green exposure using vegetation-related parameters and they were consistently correlated with improvements in self-reported general health [25, 30, 36, 46–59], and reduced mortality [60, 112–119, 121–128], as well as a decrease in the prevalence of cardiovascular and respiratory diseases [25, 72, 105, 132–134, 136, 137, 141–144], and type 2 diabetes [145–147]. Improvements in overall health can be attributed not only to the presence of green spaces in residential areas, but also to the opportunity to use and enjoy these spaces, which promotes physical activity and mitigates air pollution and urban heat. Exposure also reduces stress and increases social cohesion [25, 36, 53, 55]. However, studies that consider only grass or tree cover with less structural diversity, or the perceived quantity and/or quality of green spaces, have generally reported weaker or no associations [36, 47, 53, 54, 104]. This indicates that not all forms of greenery have equivalent health benefits. The topic of mental health and well-being has been the most extensively studied and has provided the strongest evidence: over 90% of mental health studies and 87% of well-being studies reported beneficial associations (see Tables 1 and 5). Green residential exposure, higher urban tree cover, and richer biodiversity have been consistently associated with a lower risk of depression [33, 61, 64], anxiety and stress [38, 79–81, 97–100], and greater vitality and life satisfaction [37, 76, 82, 101]. It has also been shown that mental health improves in individuals who move to greener residential areas [64]. Interestingly, perceived biodiversity and green exposure are more closely related to perceived well-being than objective biodiversity indices (e.g., NDVI, Shannon Index), highlighting the role of cognitive and emotional factors in mediating psychological recovery [76, 84, 85, 90, 97]. One of the key findings of this analysis is that biodiversity does not appear to provide any additional health benefits beyond those associated with exposure to green spaces alone. Studies suggest that commonly used indicators of green exposure are nonetheless associated with protective health effects. However, these indicators of green exposure do not reflect the richness and biodiversity of natural environments that are relevant to health (potentially leading to an incorrect or underestimated classification of exposure) [21]. Moreover, the distinction between perceived and objectively measured biodiversity is also noteworthy. Several studies indicate that perceived biodiversity may be equally or more strongly associated with well-being outcomes, highlighting the importance of individual perception and experience in mediating health benefits [20, 36]. This underscores the need for more integrative approaches that consider both ecological and psychosocial dimensions of exposure. Although limited in number, studies using bird and insect richness as an index of biodiversity have also suggested positive associations with mental health and psychological well-being [12, 26, 28, 29, 34, 37, 38]. These findings are consistent with the “biophilia hypothesis”, which posits that humans possess an innate affinity for other living organisms, and with experimental findings showing that exposure to multisensory natural stimuli, such as birdsong or insect activity, improves mood and attention recovery [164]. An emerging line of research on population mental health examines the correlation with green exposure using data on the prescription or sale of psychotropic drugs as an objective indicator of mental health [20, 45, 106–111]. All of the studies examined in this field have demonstrated an inverse relationship between exposure to green spaces (as measured by NDVI and tree cover percentage) and the use of antidepressants and anxiolytics (Table 5). These results corroborate self-reported studies indicating increased well-being and reduced stress, suggesting that urban green spaces can exert clinically significant protective effects. Observing dose-response relationships and stronger effects in socioeconomically disadvantaged populations supports the idea that equitable access to rich green spaces could help reduce mental health disparities [20, 106, 109]. There is consistent evidence of a positive association between green exposure and a reduced risk of all-cause and cause-specific mortality, particularly for cardiovascular and respiratory diseases [60, 112–128]. Several studies have demonstrated that these protective effects endure even when accounting for air pollution as a risk factor for these diseases [60, 114, 116, 118, 121, 122], as well as socioeconomic status and lifestyle factors such as smoking and alcohol consumption [60, 114, 118, 119, 121, 124, 125, 128]. This suggests that urban green exposure makes an independent positive contribution. These mechanisms may include reduced air pollutants and noise, improved microclimate regulation and increased health-promoting behaviours, such as greater physical activity and reduced stress. Evidence of a reduced risk of obesity and diabetes related to green exposure in the area of residence also supports its role in promoting healthier lifestyles and a lower prevalence of metabolic diseases. Although fewer in number, studies examining neonatal health outcomes provide suggestive evidence that maternal green exposure is associated with better birth outcomes, including higher birth weight and reduced risk of preterm birth [151, 153, 157–159, 163]. However, the strength and direction of these associations varied depending on the urban context [153] and socioeconomic status [158], and several studies reported null or conflicting results [152, 154, 156, 160, 162]. These inconsistencies may reflect heterogeneity in study design (wide variation in sample size), differences in the assessment of green exposure, and the absence of objective measurements of some environmental or social variables, with neighbourhood of residence often used as a proxy for exposure [152, 154, 157]. Nevertheless, the general trend towards positive associations supports the hypothesis that biodiverse environments may influence early childhood health outcomes by reducing stress, improving air quality, and enhancing the psychosocial well-being of mothers. One study well represents the paradigm between green exposure (maternal residence) and biodiversity (Shannon Index) and correlation with neonatal health outcome [163]: green exposure was positively correlated with birthweight especially in low-biodiversity (urban) areas, but the influence of biodiversity itself on concrete health outcomes remains poorly understood, as it has not been sufficiently investigated to date. Socioeconomic status emerged as a consistent effect modifier across all health domains. The strongest positive associations between green exposure and health were often found among individuals with lower SES [64, 67, 73, 77], older adults [79] and those with poorer baseline health [53]. This suggests that access to biodiverse green spaces may improve the health of the general population and help to mitigate social inequalities in health. The urban context also played a role: green areas that were more fragmented or distant were less effective, whereas smaller, nearby, structurally diverse green areas produced significant benefits. This highlights the importance of spatial planning and the connectivity of urban nature [47–49, 53]. The evidence shows that exposure to green environments that are rich in biodiversity, particularly those with an abundance of diverse tree cover, has a measurable positive impact on general health and mental wellbeing, and helps to prevent chronic diseases. Therefore, biodiversity is not only an ecological resource, but also a vital component of urban public health infrastructure. Integrating biodiversity enhancement into urban planning and public health policies could significantly benefit human health as well as the ecosystem. This review has several limitations. Most included studies were cross-sectional, limiting causal inference, and many relied on self-reported health outcomes, which limit accurate estimation of the association. Objective biomarkers of physiological response (e.g. cortisol, immune parameters or metabolic markers) remain underutilised [18, 40, 44]. Most green exposure assessments have focused on the quantity of vegetation rather than its ecological quality, and relatively few studies have quantified diversity at the level of plant species [12, 28, 32, 35, 36, 39, 84, 99, 151, 165]. Including fauna diversity, soil microbial exposure, and ecosystem function indicators in future research could improve our understanding of the biological mechanisms underlying the observed associations. Longitudinal and intervention studies are also needed to clarify the temporal relationships and potential threshold effects of this exposure on different health outcomes. Heterogeneity in exposure metrics and study designs also complicates comparisons across studies. Future research should consider standard and multidimensional measures not only of green exposure but mainly of biodiversity. Greater attention should also be given to the role of perceived biodiversity, accessibility, and actual use of green spaces. Integrating these dimensions may improve the understanding of the mechanisms linking not only green exposure but also biodiversity and health. From a public health perspective, these findings highlight the importance of considering not only the quantity but also the quality of urban green spaces. Biodiversity-rich environments appear to provide greater health benefits than homogeneous green areas, suggesting that urban planning and public health policies should incorporate ecological complexity as a key design principle. Enhancing biodiversity within cities may represent a cost-effective strategy to improve population health, reduce health inequalities, and promote more resilient urban environments. Conclusion This systematic review confirms that green exposure within urban spaces is positively associated with multiple adult health outcomes, with the most consistent evidence emerging for mental health, well-being, and reductions in psychotropic medication use. Socioeconomic status frequently modified associations, highlighting the importance of equitable access to green environments. Nevertheless, green exposure and biodiversity are not always directly comparable in terms of their correlation with health outcomes. The role of biodiversity in urban green spaces remains ambiguous and understudied. There is a discrepancy between perceived and measured biodiversity when specific indicators are used. Future studies should incorporate more comparable parameters of biodiversity, include longitudinal or intervention designs, and evaluate potential dose–response relationships because the impact of biodiversity on health outcomes is not yet fully understood. Greater attention should be given to the role of perceived biodiversity, accessibility, and actual use of green spaces. Integrating these dimensions into urban planning and public health policy may strengthen the health-promoting potential of urban nature. Supplementary Information Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2. Supplementary Material 3. Supplementary Material 3.