Nature’s mental health benefits may differ depending on how nature is defined and measuredAdolescent mental health in the United States is facing a growing crisis. Although an expanding body of research suggests that contact with natural environments may benefit mental health, these benefits can vary depending on how “nature” is defined and measured, underscoring the need for standardized reporting guidelines in nature–health research. Additionally, existing evidence is largely based in European contexts, highlighting the need for research in nationally representative U.S. samples, particularly amid declining adolescent mental health. To address these gaps, the present study used data from the Adolescent Brain Cognitive Development (ABCD) Study to examine the relationship between nature exposure and mental health in a large, geographically diverse U.S. sample. The study also applied the Preferred Reporting Items in Greenspace Health Research (PRIGSHARE) guidelines to evaluate the relevance, quality, and accuracy of commonly used nature metrics and identify those best suited to epidemiological research on mental health.
The study examined baseline ABCD data collected from 2016 to 2018 on 11,865 children (age 9-10) from 21 research sites across the U.S. The Child Behavior Checklist (CBCL) was completed by caregivers to assess children’s behavioral and emotional difficulties. Responses were derived into three overarching scores: internalizing problems (e.g., depression, anxiety), externalizing problems (e.g., aggressive behaviors, conduct disorder), and total problems (e.g., social problems, thought problems, attention problems). To examine specific mental health symptom domains, 17 CBCL subscales were also assessed. These included: (1) syndrome subscales for anxious/depressed, depressed, somatic complaints, social problems, thought problems, attention problems, rule-breaking behavior and aggressive behavior; (2) mental disorders-oriented scales for affective problems, anxiety problems, somatic problems, attention deficit hyperactivity disorder (ADHD), oppositional defiant problems (ODD), and conduct problems; and (3) additional scales for cognitive tempo, obsessive-compulsive disorder (OCD), and stress problems. To evaluate each participant's nature exposure at the residential census tract level, five measures of nature were considered, including: (1) percent tree canopy cover, (2) percent park cover, (3) percent natural land cover, (4) “greenness” (normalized difference vegetation index), and (5) open water (normalized difference water index). The suitability of each measure for an epidemiological study of mental health was systematically evaluated using PRIGSHARE, a reporting framework developed to help researchers assess nature measures for studies on nature and health. The evaluation focused on the measures' relevance (established associations with mental health), quality (minimal classification bias), and accuracy (finer spatial resolution). After identifying appropriate nature measures for the study, statistical analyses were conducted to examine the relationship between nature exposure and mental health, while controlling for age and gender. Analysis also accounted for family factors (family mental illness, financial history, and income) and neighborhood factors (perceived safety, crime, neighborhood deprivation, and population density) that may influence the relationship between nature exposure and mental health.
The first phase of the study asked, “What measures of nature are relevant, of suitable quality, and reasonably accurate for an adolescent mental health study?” Based on the PRIGSHARE framework, only two nature measures, percent park and percent tree-canopy cover, were identified as appropriate metrics for this study. Both measures were deemed relevant based on prior research linking park and canopy cover to mental health outcomes in adolescent populations. Both measures met quality criteria, with data available from the ABCD recruitment window (2016–2018) and spatial resolution appropriate for modeling environmental exposure. Both measures also met accuracy criteria. Park data were derived from municipal and national sources and supplemented with remote sensing and other data, such as Google Street View. Percent tree canopy cover was considered accurate because it provided a spatially continuous, remotely sensed measure of vegetative structure across geographic areas. While both greenness (NDVI) and open water met the relevance criteria, they did not meet the quality or accuracy criteria. National land cover data did not meet the criteria for relevance, quality, or accuracy.
The second phase of the study asked, “Is more available nature associated with better adolescent mental health?” Results indicated that adolescents living in census tracts with greater tree canopy cover had modestly better mental health, including fewer internalizing and externalizing symptoms. However, tree canopy cover was not related to total emotional or behavioral problems. After accounting for family and neighborhood factors, only the relationship between tree canopy cover and reduced internalizing symptoms remained statistically significant. Percent park cover was linked to modestly better mental health, but no significant relationships were detected between park cover and internalizing, externalizing, or total emotional or behavioral problems. Examination of specific symptom domains on the 17 CBCL subscales revealed significant beneficial relationships between tree canopy cover and several mental health outcomes within the internalizing domain. Specifically, adolescents in areas with more tree-canopy cover exhibited fewer symptoms of anxiety/depression, somatic complaints, and depression. However, these relationships were not statistically significant in further analyses that accounted for family and neighborhood factors. Finally, surprisingly, adolescents living in areas with greater tree-canopy cover tended to have more attention problems and ADHD symptoms, and this remained statistically significant in all analyses.
Contrary to the researchers’ hypotheses, the study “found little consistent evidence that nature (e.g., trees and parks) is protective of adolescent mental health.” After adjusting for age, gender, and family and neighborhood factors, the only significant finding suggesting a protective effect was a modest relationship between greater tree canopy cover and fewer internalizing symptoms. However, in contrast to prior research, greater tree canopy cover was also linked to increased attention problems and ADHD symptoms. This inconsistency may be due to limitations in the nature metrics used in this study, which overlook details that may be important for ADHD, such as quality, accessibility, and actual use of nature. Similarly, the lack of significant findings related to park cover may also be explained by the fact that measures of park availability do not reflect actual park use. Overall, the researchers concluded that “nature can modestly benefit adolescent mental health, though the extent of these benefits may vary depending on how nature is operationalized and measured.” Future research should examine the long-term relationship between nature exposure and mental health, and might utilize measures that capture nature access, quality, use, and individual-level experiences. Nevertheless, expanding access to natural spaces and opportunities to engage with nature may represent a promising avenue for supporting adolescent mental health in the U.S.
The Bottom Line