Department of Forest and Conservation Sciences, Faculty of Forestry, University of British Columbia, 2424 Main Mall, Vancouver, BC, V6T 1Z4, Canada.
Human Early Learning Partnership, School of Population and Public Health, University of British Columbia, 2206 East Mall, Vancouver, BC, V6T 1Z3, Canada.
Int J Hyg Environ Health. 2021 Jun;235:113767. doi: 10.1016/j.ijheh.2021.113767. Epub 2021 May 11.
Several studies have assessed the relationship between exposure to natural environments (NEs) and childhood mental health and development. In most cases, a positive association has been found, but results are inconsistent, and the strength of association is unclear. This inconsistency may reflect the heterogeneity in measurements used to assess NE.
This systematic review aims to identify the most common NE metrics used in childhood mental health and development research. Our second aim is to identify the metrics that are most consistently associated with health and assess the relative strength of association depending on type of NE exposure measurement, in terms of metric used (i.e., measurement technique, such as remote sensing), but also rate (i.e., spatial and temporal exposure).
We used the PRISMA protocol to identify eligible studies, following a set of pre-defined inclusion criteria based on the PECOS strategy. A number of keywords were used for retrieving relevant articles from Medline, Embase, PsychINFO, and Web of Science databases between January 2000-November 2020. From these, we extracted data on type of NE measurement and relative association to a number of indicators of childhood mental health and development. We conducted a systematic assessment of quality and risk of bias in the included articles to evaluate the level of evidence. Case studies and qualitative studies were excluded.
After screening of title (283 studies included), abstract, and full article, 45 studies were included in our review. A majority of which were conducted in North America and Europe (n = 36; 80%). The majority of studies used land use or land covers (LULC, n = 24; 35%) to determine exposures to NEs. Other metrics included the normalized difference vegetation index (NDVI), expert measures (e.g., surveys of data collection done by experts), surveys (e.g., self-reported assessments), and use of NE (e.g., measures of a participant's use of NE such as through GPS tracts or parent reports). Rate was most commonly determined by buffer zones around residential addresses or postal codes. The most consistent association to health outcomes was found for buffers of 100 m, 250 m, 500 m, and within polygons boundaries (e.g., census tracts). Six health categories, academic achievement, prevalence of doctor diagnosed disorders, emotional and behavioral functioning, well-being, social functioning, and cognitive skills, were created post hoc. We found sufficient evidence between NDVI (Landsat) and emotional and behavioral well-being. Additionally, we found limited evidence between LULC datasets and academic achievement; use of NE, parent/guardian reported greenness, and expert measures of greenness and emotional and behavioral functioning; and use of NE and social functioning.
This review demonstrates that several NE measurements must be evaluated further before sufficient evidence for a potential association between distinct NE exposure metrics and childhood mental health and development can be established. Further, we suggest increased coordination between research efforts, for example, by replication of studies and comparing different NE measurements systematically, so that effect sizes can be confirmed for various health outcomes. Finally, we recommend implementing research designs that assess underlying pathways of nature-health relations and utilize measurement techniques that adequately assess exposure, access, use, and perception of NEs in order to contribute to a better understanding of health impacts of surrounding natural environments.
多项研究评估了暴露于自然环境 (NEs) 与儿童心理健康和发育之间的关系。在大多数情况下，都发现了积极的关联，但结果不一致，关联的强度也不清楚。这种不一致可能反映了用于评估 NE 的测量方法的异质性。
本系统评价旨在确定用于儿童心理健康和发育研究的最常见的 NE 指标。我们的第二个目标是确定与健康最一致相关的指标，并根据 NE 暴露测量的类型评估关联的相对强度，包括使用的指标（即遥感等测量技术），以及速度（即空间和时间暴露）。
我们使用 PRISMA 方案根据基于 PECOS 策略的一套既定纳入标准确定合格研究。从 Medline、Embase、PsychINFO 和 Web of Science 数据库中使用了一些关键词来检索 2000 年 1 月至 2020 年 11 月期间的相关文章。从中，我们提取了关于 NE 测量类型和与儿童心理健康和发育的一些指标的相对关联的数据。我们对纳入的文章进行了系统的质量评估和偏倚风险评估，以评估证据水平。排除了案例研究和定性研究。
在筛选标题（包括 283 篇文章）、摘要和全文后，共有 45 篇文章纳入我们的综述。其中大多数（n=36；80%）是在北美和欧洲进行的。大多数研究使用土地利用或土地覆盖物 (LULC) 来确定对 NE 的暴露（n=24；35%）。其他指标包括归一化差异植被指数 (NDVI)、专家测量（例如，由专家进行的数据收集调查）、调查（例如，自我报告的评估）以及 NE 的使用（例如，通过 GPS 轨迹或父母报告来衡量参与者对 NE 的使用）。速度最常通过住宅地址或邮政编码周围的缓冲区来确定。与健康结果最一致的关联是在 100m、250m、500m 缓冲区和多边形边界内（例如，人口普查区）发现的。根据后验分析，创建了六个健康类别，包括学业成绩、医生诊断障碍的患病率、情绪和行为功能、幸福感、社会功能和认知技能。我们发现，NDVI（Landsat）与情绪和行为幸福感之间存在充分的证据。此外，我们发现 LULC 数据集与学业成绩之间存在有限的证据；使用 NE、父母/监护人报告的绿色植物覆盖率、专家测量的绿色植物覆盖率和情绪与行为功能；以及使用 NE 和社会功能之间存在有限的证据。
本综述表明，在能够确定不同 NE 暴露指标与儿童心理健康和发育之间潜在关联的充分证据之前，必须进一步评估若干 NE 测量方法。此外，我们建议加强研究工作之间的协调，例如，通过复制研究和系统地比较不同的 NE 测量方法，以确认各种健康结果的效应大小。最后，我们建议实施评估自然-健康关系潜在途径的研究设计，并利用适当评估自然环境暴露、获取、使用和感知的测量技术，以更好地了解周围自然环境对健康的影响。
