Effect of greenness on asthma in children: A systematic review

Abstract Greenness such as trees, plants, and shrubs may positively influence mental and physical health, but the relationship between greenness and asthma is poorly understood. Because asthma is the most prevalent child respiratory disease internationally, elucidating the role of greenness may substantially benefit public health. The purpose of this systematic review was to synthesize findings related to effects of greenness on asthma in children. Following PRISMA guidelines, six databases were searched for international publication of primary research results relevant to the relationship between greenness and child asthma. Of 82 initial results, seven articles remained after removal of duplicates and applying exclusion criteria. Six reported no direct association between greenness and child asthma, while one found increased greenness protective for asthma. None found a negative direct association between greenness and child asthma. Evidence supported benefits of greenness on child asthma through mediation of factors such as exposure to tobacco smoke, high traffic volume, and difficult family relationships. Even without a direct association, greenness can be considered a public health asset as it may mediate other factors contributing to asthma in children. Public health nurses can use these findings to educate clients and partners while advocating for policies to protect greenness.

and respiratory health, longevity among the elderly, cognitive restoration, increased self-discipline, reduced aggression, improved birth outcomes, reduced crime and interpersonal violence, and increased social ties and quality of life (Hartig, Mitchell, de Vries, & Frumkin, 2014;Nieuwenhuijsen et al., 2014;Villeneuve et al., 2012). Some potential pathways by which vegetation contributes to health include moderation of ambient temperature, absorption of air pollution, and providing opportunity for physical activity .
Asthma is the most prevalent child respiratory disease, affecting 235 million people worldwide (World Health Organization [WHO], 2017) and 24.6 million Americans, including 6.2 million children (CDC, 2017). In the United States, costs of healthcare, lost school time for children, and lost work time for caregivers add up to $56 billion annually (Children's Hospital Colorado, 2019). In 2013, 1.6 million visits to emergency departments in the United States were attributed to asthma exacerbation, and over 3,600 Americans died from asthma in 2015 (CDC, 2017). Because asthma is exacerbated by environmental exposures, it remains important for public health nurses to explore modifiable factors to improve outcomes for those affected, including characteristics of the physical environment such as greenness.
While the health benefits of greenness may be numerous, findings related to child asthma are mixed. A systematic review and meta-analysis by Lambert et al. (2017) examined evidence regarding the relationship between greenness and allergic respiratory disease and asthma in children. Studies included examining asthma as the outcome demonstrated inconsistent results. Sbihi, Tamburic, Koehoorn, and Brauer (2015) found an inverse relationship between greenness and asthma for preschool children but F I G U R E 1 PRISMA Flow Diagram depicting literature search results (adapted from Moher et al., 2009)  found no relationship among school-aged children. Lovasi, Quinn, Neckerman, Perzanowski, and Rundle (2008) found decreased asthma among children living in areas with more trees, but Lovasi et al. (2013) found increased greenness around the home significantly related to increased asthma at age seven. Andrusaityte et al. (2016) found greenness detrimental to child asthma, while others found no statistically significant relationship (Dadvand et al., 2014;Pilat et al., 2012). As considerations for future work, Lambert et al. (2017) discussed heterogeneity of findings and inconsistency in operational definitions of greenness, seasonal variation and proximity of greenness to be measured, and confounders such as proximity to high-traffic roads and urban versus rural areas. Metaanalysis added further to the body of evidence for discerning the relationship between greenness and child asthma, and additional studies have emerged since the review was published in 2017. The purpose of this literature review is to evaluate literature regarding greenness and asthma in children published since 2017, and to consider implications of this relationship for public health nursing.

| Design
This review was conducted following the PRISMA Statement (Moher, Liberati, Tetzlaff, & Altman, 2009 and Urban Studies Abstracts. Given the interdisciplinary science related to greenness and health, these databases were chosen for results relevant to both health and environment. The search string was developed through repeated attempts to isolate terms specific to the constructs of greenness and asthma, using MeSH headings, searching known literature for keywords, and using truncation symbol (*) to search for all versions of the root word. The final search string included: [("greenspace*" OR "greenness" OR "green space*" OR "tree canopy" OR "vegetation" OR "street tree*" OR "urban forest*") AND ("asthma*" OR "respiratory" OR "lung function") AND ("child*" OR "teen*" OR "adolescen*" OR "youth") in the abstract.
Inclusion criteria were international publications of primary research results relevant to the relationship between asthma and measured greenness in children under age 18, written in English language. International studies were included, as research related to greenness is emerging globally. Studies included in the Lambert et al.

| Analytic strategy
Data were evaluated using created tables. Based on suggestions to standardize measurement of greenness (Lambert et al., 2017), conceptual, and operational definitions of exposures and outcomes are provided in Table 1. Study objective, characteristics, methods, and findings of each study were considered separately (Table 2).

| RE SULTS
Initial search yielded 82 results. Removal of 31 duplicates and ten articles not in English language left 41 results. Title and abstract review excluded 28 articles that did not fit the research purpose: six were not primary research, 11 were not related to asthma, and 11 were unrelated to measured greenness. Although all publication dates were included in the search, no results appeared before 2017 that were not included in the review by Lambert et al. (2017).
Of 13 remaining results, six were included in the previous review and excluded here to prevent over-representation of findings.
In total, seven articles were included in this analysis, consisting of six primary research studies and one systematic review with meta-analysis.
The grading system used for this review was the Johns Hopkins

Research Evidence Appraisal tool, part of the Johns Hopkins
Nursing Evidence-Based Practice model. The goal of the model is to ensure research findings and best practices are quickly and appropriately incorporated into patient care (Dang & Dearholt, 2017). Because all included articles were observational, all were graded at level III.

| Terminology
Several terms are used to describe the concept of greenness, including "greenness," "green space," "greenspace," and "vegetation." Frequency table of terms used in included studies is provided in Table 3. combined NDVI with land-use classification as another measure of greenness: total and native land-use types (Donovan et al., 2018) and urban green areas (Tischer et al., 2017). Feng and Astell-Burt (2017) did not use NDVI but used land classified as "parkland".

| Measurement of greenness
Average NDVI is calculated within a circular radius (buffer) from a specified point, usually home address. Among the six studies using NDVI, four examined only one buffer distance: 100m (Lambert et al., 2017), 200m (Cillufo et al., 2018, 250m (Chen et al., 2017), and 300m (Tischer et al., 2017). One study used nested buffers of 100m, 250m, and 500m (Eldeirawi et al., 2019), a practice suggested for approximating greenness near the home, neighborhood, and larger community. One article did not specify buffer, but used meshblocks, statistical subdivisions not comparable to buffer radii as they are nonuniform in size (Donovan et al., 2018;Stats NZ, n.d.).

| Greenness and asthma
Out of the seven reviewed papers, six reported no statistically significant direct relationships between greenness and child asthma while one (Donovan et al., 2018) found 1 standard deviation increase in NDVI was associated with a 6% lower risk of asthma (95% CI: 1.9%-9.9%).
Three papers reported greenness was protective for child asthma via mediation of other negatively related health factors, such as difficult family relationships (Chen et al., 2017), high traffic volume (Feng & Astell-Burt, 2017), and tobacco smoke exposure (Eldeirawi et al., 2019). While Chen et al. (2017) reported no main effects of greenness on asthma control or functional limitations of children with asthma, interaction effect of greenness x parent hostility showed an inverse relationship protective for functional limitations of children with asthma. Feng and Astell-Burt (2017) reported no association between greenness and asthma incidence for children not exposed to heavy traffic, but children living in areas of both high traffic and low greenness (characterized as 0%-20% land coverage) had higher risk of asthma (OR 1.87, 95% CI: 1.37-2.55) and those living in high traffic and high greenness (>40% land coverage) had lower risk (OR 0.32, 95% CI: 0.12-0.84).

| D ISCUSS I ON
In this emerging area of scientific interest, greenness is opera-   Eldeirawi et al. (2019) suggest in their discussion that their finding related to second-hand smoke exposure may be a proxy for low socioeconomic status. Two studies (Chen et al., 2017;Cillufo et al., 2018) only included individual-level confounders in analysis, neglecting the effect of community. Because asthma is affected by both individual and community-level factors, it is important to consider influences of both categories for isolating the effect of greenness.
There is growing interest in understanding how greenness can be leveraged to support public health, but there are limited published findings. Terms used to refer to the concept of greenness vary. In these results, terms included "greenness," "green space," "greenspace," and "vegetation" (Table 3). Other terms in the study of greenness may refer to trees, such as "street trees," "urban forest," "tree cover," and "tree canopy." Such variation in terminology may compli-

| Implications for public health nursing
This review found either positive or no direct association between greenness and child asthma. There was no indication greenness is detrimental to child asthma, and it may mediate factors contributing to child asthma. Together with evidence supporting the relationship between greenness and multiple health outcomes, these findings indicate greenness can be considered a community asset. As public health nurses assess risk in geographic areas and capture population assets (ANA, 2007), greenness should be included in such assessments as communities lacking greenness may be at risk for poor health outcomes.
In order to promote greenness as a community asset, public health nurses can support programs to protect and increase greenness. Arbor Day Foundation (2019), for example, offers programs such as "Tree City USA" and "Tree Cities of the World" to encourage growth and maintenance of urban tree canopy. Using tools provided by the United States Department of Agriculture's Forest Service (2019), public health nurses can assess existing tree canopy to determine need for greenness in communities served and partner with communities to organize tree planting programs. Public health nurses can identify areas that would support gardens or expanded tree coverage, such as schools, parks, playgrounds, street medians, and parking lots. Through greening initiatives, public health nurses can provide assets to build and protect community health.
Partnering with local politicians, public health nurses can advocate for policies that expand and protect greenness.

| CON CLUS ION
Greenness has no direct effect on child asthma but may be protective via modification of individual and community-level risk factors.
Results of this updated review are consistent with a prior systematic review. Further, this review found a protective effect of greenness through moderation of factors contributing to child asthma, such as difficult family relationships, high traffic volume, and tobacco smoke exposure. Public health nurses should share these findings with individual and community partners, consider greenness a community asset for promoting health, and advocate for policies and programs to increase and protect existing greenness.