A recent case study focused on Washington State reveals that the association between residential greenspace and mental illness mortality is not uniform, but instead varies across ecological subregions. While many large-scale population studies suggest that higher levels of vegetation contribute to lower mortality rates, this research demonstrates that applying identical exposure thresholds across diverse landscapes can mask nuanced health outcomes. By analyzing death records alongside local vegetation data, investigators found that the health relevance of specific Normalized Difference Vegetation Index (NDVI) values depends heavily on the surrounding ecological subregion.
The study utilized death certificate records from Washington State for the year 2018 to construct an analytical cohort of 31,245 decedents. Of these, 233 deaths were specifically attributed to mental illness, defined by ICD-10 codes F20–F69 and F99. The objective was to determine if the relationship between greenspace exposure and mortality risk remains consistent when using context-specific definitions of greenness, rather than relying on statewide absolute measurements.
Residential greenspace exposure was calculated as the five-year mean NDVI within a 15-minute pedestrian walkshed around each subject’s place of residence. To assess the impact of different exposure levels, the researchers established binary treatment indicators based on the 50th, 60th, 70th, 80th, and 90th percentiles of the NDVI distribution within each distinct ecological subregion identified in the state. This approach allowed for a more localized assessment of how different levels of greenery relate to health outcomes.
To estimate average treatment effects, the researchers employed covariate-balancing propensity score matching followed by logistic regression.
Statewide analysis indicated that treatment effect estimates varied across different NDVI thresholds without demonstrating a consistent protective association. When investigators applied absolute statewide NDVI cutoffs, the results failed to provide a stable or clear link between vegetation levels and reduced mortality. This lack of consistency across the state highlights the limitations of treating diverse ecological environments as a monolithic landscape in environmental health research.
In contrast, subregion-specific analyses revealed substantial heterogeneity in both the direction and magnitude of the estimated effects. The findings indicate that when exposure thresholds were defined according to local ecological settings, the resulting estimates were more stable and interpretable. This suggests that the health relevance of a given NDVI value is not universal, but rather contingent upon the specific ecological characteristics of the residential area.
These findings suggest that ecological stratification is an essential consideration when conducting environmental health research. The study supports the use of context-specific exposure definitions to accurately evaluate the health impacts of residential greenspace. By accounting for the diversity of local landscapes, researchers can derive more precise associations between environmental factors and mental health outcomes.
The researchers noted that larger longitudinal studies with improved exposure assessment are necessary to further clarify these complex relationships.
Subject of Research: Environmental epidemiology and mental health mortality
Article Title: Ecological context modifies estimated effects of residential greenspace on mental illness mortality: a case study of Washington State, USA
Article References: Song, I., & Luan, H. (2026). Ecological context modifies estimated effects of residential greenspace on mental illness mortality: a case study of Washington State, USA. International Journal of Health Geographics. https://doi.org/10.1186/s12942-026-00497-z
Image Credits: AI Generated
DOI: 10.1186/s12942-026-00497-z
Keywords: Residential Greenspace, NDVI, Mental Illness Mortality, Environmental Epidemiology, Washington State, Ecological Context, NDVI Thresholds, Public Health, Propensity Score Matching, ICD-10, Health Geography, Vegetation Index, Mortality Risk, Environmental Health Research, Residential Exposure
News Source: Phoebe Ingram. (November 17, 2026). Local Ecological Context Modifies Estimated Greenspace Effects on Mental Illness Mortality. Scienmag.



