Neighborhood greenspace visits and mental health: insights from mobility data across nine U.S. metropolitan areas.

In pooled models across nine MSAs, more greenspace visits were associated with better mental health after adjusting for NDVI, demographics, and social vulnerability.

• The pooled OLS model yielded β = -0.029 for greenspace visits and poor mental health prevalence.
• The pooled model R² = 0.62, indicating substantial explanatory power.
• The sample included 13,152 census tracts across nine MSAs representing 72.5 million residents.
• Models adjusted for NDVI, Social Vulnerability Index (SVI) themes, and demographics.

Metro-specific OLS models showed the strongest associations between greenspace visits and lower poor mental health prevalence in Houston and Atlanta.

• Houston showed the strongest effect: β = -0.113, R² = 0.716.
• Atlanta showed the second strongest effect: β = -0.096, R² = 0.634.
• Weaker associations were found in Philadelphia and Washington DC.
• Poor mental health prevalence ranged from 12.9% in DC to 18.7% in Houston across the nine MSAs.

Greenspace visit rates varied substantially across metropolitan areas, with DC having the highest and Houston the lowest visit rates.

• Greenspace visits ranged from 11.5 (Houston) to 33.2 (DC) across the nine MSAs.
• Poor mental health prevalence ranged from 12.9% (DC) to 18.7% (Houston).
• DC had both the highest visit rate and the lowest poor mental health prevalence.
• Houston had both the lowest visit rate and the highest poor mental health prevalence.

Multi-scale geographically weighted regression (MGWR) revealed spatial heterogeneity in the association between greenspace visits and mental health within metropolitan areas.

• MGWR revealed strong urban-core associations in Houston, Atlanta, and Dallas.
• Peripheral tracts often showed null effects in these same metros.
• Spatial heterogeneity was assessed using MGWR to capture variation at multiple geographic scales.
• The spatial analysis demonstrated that neighborhood-level associations varied considerably within city boundaries and suburban areas.

The study integrated multiple data sources to examine neighborhood-level greenspace visits and poor mental health across nine U.S. metropolitan statistical areas.

• Data sources included greenspace visits, CDC PLACES mental health estimates (2023), Social Vulnerability Index (2018), ACS demographics (2010–2014), and NDVI (2021).
• The unit of analysis was the census tract level across nine MSAs.
• Both ordinary least squares (OLS) and multi-scale geographically weighted regression (MGWR) models were employed.
• The study distinguished between greenspace presence (measured by NDVI) and actual greenspace use (measured by visits).

Most prior studies have focused on greenspace presence rather than actual use, and fewer have explored how associations vary within city boundaries and suburban areas.

• The authors note that access to greenspace supports mental health but most studies focus on the presence of greenness rather than actual use.
• The study was designed to address the gap in understanding how greenspace engagement (visits) relates to mental health independent of greenspace presence (NDVI).
• The study examined spatial heterogeneity within metropolitan areas, a dimension underexplored in prior literature.