Associations between neighborhood greenspaces and cognitive and brain volume measures in cognitively normal older adults

Abstract

AbstractBackgroundLiving in neighborhoods with greater vegetation and park space has been associated with better brain health in older adults. While mechanisms relating greenspaces and brain health are not well understood, greenspaces may lead to better cognition and lower Alzheimer’s risk by providing social and physical activity opportunities and stress reduction, factors previously associated with Alzheimer’s disease. This study investigated associations between neighborhood greenspace types and multiple brain health measures for Alzheimer’s disease and related disorders, which has been examined little to date.MethodResidential addresses and magnetic resonance imaging (MRI) volumes were obtained from cognitively‐normal participants who completed Uniform Data Set assessments from Knight Alzheimer’s Disease Research Center (St. Louis, Missouri). Addresses geocoded to US Census tracts were linked to National Land Cover Dataset data to identify open spaces and areas with tree canopy (hereafter termed forest). Longitudinal clinical data were obtained from the National Alzheimer’s Coordinating Center. Outcomes included CDR® Sum of Boxes, four cognitive domain z‐scores (memory, attention, executive function, and language) constructed from ten neuropsychological tests, and hippocampal and white matter hypointensity volumes (167 had T1‐weighted MRI). Multi‐level linear regression models (clustering on Census tract and participant) controlled for demographics, comorbidities, and neighborhood disadvantage and population density. Brain volume models additionally controlled for intracranial volume.ResultParticipants (n=553) were on average 74 years old (standard deviation [SD]=7), 16% were Black, and 81% were White. Neighborhood percentage of open space ranged from 0 to 42% (mean=6.6%, SD=9.9) and forest ranged from 0‐16% (mean=0.2%, SD=1.4). Greater percentage of forest was associated with lower baseline episodic memory (estimate=‐0.79; 95% confidence interval [CI]=‐1.13, ‐0.46), slower annual episodic memory declines (estimate=0.069, 95% CI=0.002, 0.137), higher baseline executive function (0.15, 95% CI=0.03, 0.26), and fewer white matter hypointensities (estimate=‐3,521 mm3, 95% CI=‐6,443, ‐600). No other associations were observed.ConclusionLiving in neighborhoods with more tree canopy was associated with lower episodic memory but higher executive function at baseline, slower annual declines in executive function, and fewer white matter hypointensities. This study suggests mixed associations between neighborhood tree canopy and brain health that will need to be replicated in larger, diverse cohorts.