Abstract

Background: Greenspace has been shown to be positively associated with mental wellbeing, but studies from the global South have been scarce. We advance the understanding of the relationship between greenspace and depressive symptoms by using multiple clearly defined metrics describing neighborhood greenness and urban parks in an understudied region with rapid urban growth. Methods: We linked individual-level health survey data for urban residents in Mexico (n = 17,258 respondents in 84 cities) with measures of greenspace such as satellite-derived normalized difference vegetation index (NDVI), percent green area, urban parks characteristics, and kernel-density-derived continuous greenspace indicator. We estimated the odds ratios of experiencing depressive symptoms associated with greenspace at residential neighborhoods adjusted for individual and area-level characteristics. Results: Among the various measures of greenspace investigated, the amount of greenness measured by neighborhood NDVI was associated with smaller odds of depressive symptoms. An increase by one standard deviation in the median of annual maximum NDVI at neighborhood level is associated with 8.7 % (Odds Ratio [OR] = 0.913, 95 % CI 0.853-0.977) lower odds of experiencing depressive symptoms, adjusted for individual and arealevel characteristics. We find some evidence that, when neighborhood-level greenness is accounted for, the broader availability of greenspace outside of the neighborhood may be associated with smaller odds of depressive symptoms. We found no statistically significant associations for the measures describing % greenspace in a neighborhood and urban parks, and the results were not sensitive to seasonal changes in greenness. Neighborhood-level particulate matter (PM2.5) may lessen the benefits of greenspace for depressive symptoms. Conclusions: Higher neighborhood-level greenness as measured by NDVI is associated with smaller odds of depressive symptoms in Mexico's cities, whereas many other metrics are not. The influence of the choice of greenspace metrics on the subsequent associations highlights the importance of clear operational definitions of greenspace and the need to consider multiple complementary greenspace metrics in greenspace-health research.