Street view-based exposure to greenspace and mortality: the Multi-Ethnic Study of Atherosclerosis

Higher exposure to greenspace has been associated with reduced mortality rates in predominantly White samples. However, the specific contribution of different types of greenspace exposures to these associations and whether the associations vary by race and ethnicity have not been studied. Our main objective was to quantify the association between specific types of greenspaces and all-cause mortality risk among participants of the Multi-Ethnic Study of Atherosclerosis (MESA). We analyzed data from 6,795 MESA participants, with baseline interviews in 2000 and follow-up through 2018. Deep learning algorithms were applied to Street View images to estimate percent of residential greenspace type (e.g., %Trees and %Grass) within a 500 m buffer for each follow-up year. We used Cox models to estimate hazard ratios (HR) for the association between % of each greenspace metric and mortality, adjusting for demographic and socioeconomic factors. Among 6,748 individuals (mean baseline age 62 years [SD 10]) totaling 973,608 person-years, we observed 1898 (28.1%) deaths during 19 years of follow-up. In fully-adjusted models, exposure to %Other green (fields, flowers, and plants) was associated with lower mortality (0.89, 95% CI: 0.84, 0.94), while %Grass showed a similar but less consistent inverse association (0.91, 95% CI: 0.83, 0.99). This protective association was particularly notable among participants living in neighborhoods with low socioeconomic status (%Other green 0.86, 95% CI: 0.77, 0.97). No associations were observed for %Trees. Our results suggest that greater exposure to certain greenspace types may be associated with a reduced mortality risk among MESA participants.