Green Space Exposure and Cognition: The Multi-Ethnic Study of Atherosclerosis

Abstract

RATIONALE: The effects of long term green space exposure on cognition are not fully understood. With the aging of the world’s population, increasing prevalence of Alzheimer’s Dementia, and continued detachment from nature, we seek to contribute to quality research investigating the association of long term green space exposure with cognition. OBJECTIVES: To determine if long term exposure to green space is associated with better cognition, measured by Normalized Difference Vegetation Index (NDVI) and Cognitive Abilities Screening Instrument, Version 2 (CASI), respectively. METHODS: We performed analyses of participants enrolled in the MESA (Multi-Ethnic Study of Atherosclerosis) Exam 5 administered from April 2010 through February 2012. This population-based cohort was aged 45–84 years during the initial MESA study July 2000-August 2002. Time-weighted averages of NDVI in 250m and 500m buffers were calculated at individual addresses for the preceding ten years for 4,106 participants. CASI measures taken during Exam 5 were available for 4,278 participants. Regression models were used to determine whether CASI scores were associated with NDVI. Models were adjusted for age, race/ethnicity, income, education, predicted exposure to particulate matter of 2.5 micron aerodynamic diameter or less (PM2.5), apolipoprotein E allele polymorphism status and location (site). Additionally, regression models were used to determine if the presence of certain apolipoprotein E allele polymorphisms modify the association. RESULTS: NDVI score increment of one standard deviation (SD), 33.55 points, was associated with 0.47 (95% CI .007, .021, p<0.001) greater CASI score, controlling for potential confounders including age, race/ethnicity, education, income, apolipoprotein E allele status, and predicted air pollution (PM2.5). With the addition of site into the model, the association of NDVI and CASI became statistically non-significant, with an effect estimate of 0.30 (95% CI -0.004, 0.022, p=.203). The apolipoprotein E4 allele modified the association between NDVI and CASI, with strongest effects found in those with two copies of the apolipoprotein E4 allele. CONCLUSIONS: Increased proximal (250-500m) residential greenspace, as measured by NDVI, was positively associated with better performance on cognitive testing validated for dementia screening, as measured by CASI, in an elderly population (mean age 69.7 y). The association was strongest among those with two copies of the apolipoprotein E4 allele, a polymorphism associated with higher rates of developing Alzheimer’s disease dementia. Limiting the NDVI comparisons to within-city contrasts resulted in attenuation of the association.